Final Report Summary - THE ONE STUDY (A Unified Approach to Evaluating Cellular Immunotherapy in Solid Organ Transplantation)
The success rates of transplant surgery have improved remarkably over the last half century, making this procedure a life-saving option for many patients with organ failure. Early outcomes in transplant recipients are outstanding and patients normally recover from surgery with a well-functioning replacement organ and typically return to an active lifestyle in a short period.
Unfortunately, transplant practice is complicated by the fact that the adult human immune system is strongly biased towards damaging reactions against allogeneic tissues, resulting in total donor organ destruction within a matter of weeks after transplantation, unless the immune system is profoundly inhibited. To impede the immunological response, researchers have developed an armamentarium of general immunosuppressive drugs necessary for sparing transplanted organs from early destruction. During the therapy with immunosuppressive drugs the whole immune system is impaired and a myriad of side-effects arise. Thus, patients maintained on conventional immunosuppressive treatment suffer the consequences of drug toxicity, the development of chronic rejection, reduced resistance to infections, and a high rate of cancer occurrence. Besides these important side effects, financial costs can be high for the families and for health care systems. Added to the reality of presently available treatment options is the fact that 10-year organ survival rates in renal transplantation have astonishingly not shown improvement over the last decades. Improvements in treatment for these patients is imperative.
The ONE Study Focus
Preventing immunological rejection of transplanted organs with less need for long-term use of pharmacological immunosuppression is a primary objective. New transplant research should concentrate on early strategies that support long-term immunological acceptance of transplants, allowing for at least a reduction in the use of general immunosuppression. It would dramatically improve the outcome for transplant recipients and reduce healthcare costs. A means to achieve this goal has not been realized with pharmacological or biological agents, so we must now look towards new, innovative approaches.
The ONE Study applies the novel concept of cell therapy to human clinical organ transplantation. This cooperative project was aimed at developing and trialling various immunoregulatory cell products in organ transplantation recipients, allowing a direct comparison of the safety, clinical practicality and therapeutic efficacy of each cell type.
The central focus of The ONE Study project was to:
▪ Produce and manufacture distinct population of haematopoietic immunoregulatory cells
▪ Comparatively study the tolerogenic characteristics of these regulatory cell types
▪ Test these cell therapy products side by side in a clinical trial of living donor renal transplant recipients
Project Context and Objectives:
The principal objective of The ONE Study was to establish whether purified haematopoietic regulatory cells can be used therapeutically to modulate the immunological response of recipients of transplanted organs, thus reducing the level of pharmacological immunosuppression needed.
To assess this, several regulatory cell products were licensed for manufacture under good manufacturing practice (GMP) conditions and were tested in a coordinated clinical study. By directly evaluating the various immunoregulatory cell therapies against one another in this way, and also against a group of patients not receiving cell therapy, The ONE Study provided a guideline for the scientific and medical community.
We have chosen to include in The ONE Study those cell preparations which, represent the most promising therapeutic agents and which are closest to clinical application:
Six cell products have been approved by national and local authorities for use in the ONE Study clinical trials. Four of these cell products were developed in Europe and two in the United States.
The products developed and the date of cell product approval for use in The ONE Study is given below:
Country Partner Cell Product Name Date of Approval
Germany UHREG M reg_UKR April 2014
Germany Charité nTregs January 2015
France Nantes ATDC September 2014
UK Oxford/ KCL ARTreg product September 2013
USA UCSF darTregs September 2014
USA MGH Treg with belatacept July 2014
ATDC - autologous tolerogenic dendritic cells
ARTreg - autologous regulatory Treg product
darTreg - donor alloantigen reactive Tregs
Treg with belatacept - Tregs generated in the presence of donor cells and belatacept
Our partner in Milan has continued to seek approval for their cell product and for their cell therapy trial, but they have not yet started a cell therapy trial with their Tr1 cell product. The product they are developing has a complicated process that has required extra time to validate.
Determining the fate of therapeutic cells after administration to patients is central to understanding their potential immunomodulatory effects in vivo. In the context of regulatory cell populations administered by intravenous infusion, subsequently detecting the cells presents a number of technical challenges.
An innovative cell tracking technology has been developed to assess where and for how long cells traffic when administered to humans. (WP2)
A prototype ablation chamber, designed and developed by the University of Loughborough (LU), has undergone testing and has been shown to provide a 50-fold increase in speed and a six-fold increase in sensitivity for the analysis of Gd labelled T cells.
ESI’s efforts have focussed on translating the results from LU's extensive testing into a product for cell tracking. Development focussed on four key areas: spot size/resolution; laser wavelength/pulse rate; high-resolution sample viewing; sample transport speed. Tests showed that cells were typically in the range of 5-20 microns in diameter depending on cell type, however, testing showed that to adequately distinguish neighbouring cells in tissue sections or cytospins a higher resolution is required. Therefore, a laser beam delivery system was developed that could achieve a spot size range of sub-micron to 50 micron per ablation. This was challenging as it approached the diffraction limit for light passing through an aperture.
ESI developed a laser in-house to be 100% fit for purpose for this application. The Polaris266 is capable of firing 100 shots per second (100Hz), which fulfils the requirement to process 100 cells per second. The laser operates at 266nm (frequency quadrupled from a Nd:YAG diode-pumped source) for best coupling with organic material while maintaining integrity of the glass substrate.
ESI have significantly developed their optical system in order to achieve two goals: high-quality delivery of a UV beam to the sample surface, and high resolution optical viewing of the sample. A novel objective lens array has been designed and implemented into the platform to achieve this.
Sample transport technology (DCI and sniffer) developed by University of Loughborough has been integrated into the ESI laser platform in order to achieve the goal of analysing 100 cells per second.
The above developments have been incorporated into two alpha units, one of which is currently used for applications development at BAM (Federal Institute for Materials Research and Testing), Berlin, and the other will be delivered to LU. LU is testing the system and apply it to cell analysis.
Therefore, the technology is indeed very promising for the original purpose of detecting labelled cells after cell therapy, and we remain confident that this will result in broadly applicable new instrumentation with a high potential impact for eventually tracking cell in humans.
The main initial clinical activities of the centres involved in The ONE Study was to prepare the clinical protocols for the Reference Group Trial (RGT) and the Cell Therapy Group (CTG) trials, submit IMPDs (Investigational Medicinal Product Dossier) and INDs (Investigational New Drug) to the regulatory agencies for the cell therapy trials, and the development of a centralized immune monitoring program.
The ONE Study involved the design, approval and performance of 7 individual clinical trials.
The Reference Group Trial (RGT) was planned during the first year of The ONE Study funding period. This trial serves as a control or reference group for the 6 Cell Therapy Group (CTG) trials that were not planned to start before the 4th funding year. The RGT design was agreed upon by all 8 transplant centers involved in The ONE Study, and received ethical and regulatory approval by all involved authorities between September 2012 and September 2013. Patient recruitment was completed in September 2014, with the aim of recruiting 60 informative patients. In total, 70 patients were enrolled into the RGT, with 9 patients dropping out prematurely; therefore, 61 informative patients were transplanted and followed-up until the final trial visit (i.e. 60 weeks post-transplantation). The final RGT study visit (Last Patient Last Visit) took place in December 2015.
An extensive effort was made to create a portfolio of standardized & reproducible molecular assessments, multi-parametric flow cytometry as well functional assays which are now used by many laboratories world-wide. Notably, the Charité group in cooperation with Beckman Coulter has developed 6 flow cytometry panels and standardised whole blood staining procedures for our study and for commercial exploitation. Our approach on standardizing flow cytometry and resulting SOPs have been published in Transplantation Research (Streitz et al. 2013) and Methods Mol Biol (Schlickeiser et al. 2016). These antibody panels have been published and are now available commercially through Beckman Coulter. Using these established flow cytometry antibody panels we have also generated and published age- and gender-dependent differences in leukocyte subset composition (Kverneland et al. 2016). We have also completed the performances of all immune monitoring assays (flow cytometry, qRT-PCR, TSDR analysis, IFNg Elispot, CD154/CD137 assay) for all available RGT samples.
The Cell Therapy Trials (CTG). There are 6 CTG trials that were initiated within The ONE Study. Two trials tested the use of polyclonal T regulatory (Treg) cells (either frozen or fresh cells), two trials use donor-reactive T regulatory cells produced by two completely different methods, and two trials use blood monocytes as starting material to produce either tolerogenic dendritic cells (tolDC) or regulatory macrophages (Mreg).
All 6 CTG trials were initiated sometime within approximately one year after April 2014. All trials followed the same clinical trial protocol (tacrolimus+MMF maintenance, +3 months of steroid treatment), with the exception of delivery of a different cell therapy product. Also, different cell numbers were delivered in each trial and the timing of delivery relative to kidney transplantation was specific for each trial. A 7th trial was planned by our partner in Milan, where they would use a cell product enriched for Tr1 cells (another donor-reactive T regulatory cell product), but this trial was not initiated because of issues related to obtaining the approval for the trial/cell product manufacturing; this trial has not yet started and will not be part of The ONE Study group of cell therapy trials. Planned patient numbers to receive cell therapy for each cell product varied between 6 and 12. At the writing of this report, 5 of the 6 trials have either reached completion or have been stopped for different reasons explained elsewhere in this report.
The RGT has been completed, the last follow-up visit occurred in December 2015. A tailored, very complex, eCRF system has been developed by a project beneficiary and collects data from both the RGT and CTG trials.
Trial performance conclusions:
Regarding the RGT, 70 patients were recruited and this trial was completed as planned.
For the CTG trials, 60 patients were recruited into the various trials, with 38 of these patients being treated with one of the 6 cell therapy products.
All of these trials are either completed with recruitment, or have been stopped due to the reasons described in the report. As a lesson from The ONE Study, excellent recruitment in the UK by the combined Oxford and London partners bodes well for sharing efforts towards patient recruitment. Importantly, the framework for The ONE Study worked perfectly in terms of supporting the initiation of the trials. Groups freely shared their IMPD/IND documents with each other to get approval from their authorities and all groups have realized that performing these trials would have been impossible in this time frame without the support of the whole ONE Study consortium. In group meetings, we estimate that it would have taken approximately 3 more years to start and complete trials without the cooperation of our partners. Indeed, some of the groups believe they would not have even gone to the level of clinical trial performance without The ONE Study support. Therefore, the structures and mutual support provided within this consortium has opened the way to the performance of new trials in organ transplantation that would not otherwise have been possible.
“The ONE Study model” is being copied and promoted by other funding organizations (e.g. NIH) to organize new efforts in transplantation and cell therapy.
Trial evaluation. The primary objective of The ONE Study was to test the comparative safety of various T cell and monocyte derived immunoregulatory cell therapies in renal transplant recipients. The strategy for properly testing safety was to test the cell products using the same basic clinical trial protocol, so that any deleterious effects could be equally assessed. The safety assessment was expected to be one of the most critical outputs from The ONE Study, so that new trials could then be based on cell products that already have an existing safety profile.
The ONE Study has demonstrated that immune cell therapy with the tested T cell and monocyte derived cell products is safe in terms of side effects. Two cell products did raise cautionary flags, but have not been ruled to be unsafe; further testing in patients is needed to reach a conclusion regarding safety of these products. Moreover, we now have the first evidence that in fact reduction of general immunosuppression via application of cell therapy could be effective in reducing side effects associated with standard pharmacological drugs. These results have given impetus to the further development of these cell therapies for organ transplantation, which is evidenced by the approval of new cell therapy trials both in Europe and in the USA.
Although the main goal was to determine safety, we also designed the study to get at least an initial set of preliminary data regarding effects of cell therapy on development of transplant rejection and on the immune system.
The primary endpoint for the RGT was biopsy confirmed acute rejection (BCAR). Eight patients experienced a BCAR, and these results have been confirmed by the central pathologist (Prof. Ian Roberts) in Oxford. There was an uneven distribution of BCAR at the different sites, but this was not entirely unexpected since the development of BCAR is normally only in 1 or 2 patients out of 10. Nonetheless, there were 3 rejection episodes in the 10 patients in Berlin. We are not sure the reason for this high rejection rate. Based on the literature and our statistical calculations, we expected a rate of BCAR between 2.4-17.6% in the RGT. Therefore, the rejection rate was within our expectations and provides a very useful control for the CTG trials. Regarding the immune monitoring, patient samples in the RGT have been received by the central laboratory at the Charité and there are some key findings to summarize from the RGT: When combining the findings of all RGT patients we did identify a set of biomarkers (e.g. Nav3 gene expression), which allows pre-transplant identification of patients experiencing acute rejection. In addition, we detected alterations in peripheral blood immune cell composition (e.g. % of total B cells, naïve CD4+ T cells) in patients showing a decline in graft function up to 12 months prior to clinical manifestation.
Cell Therapy Group trial Primary Endpoint results: The BCAR results for the CTG trials have not all been scored yet by the central pathologist (some only scored by the local pathologist), and therefore must be considered preliminary.
The most promising trial results came from the UK polyclonal Treg trial, where they treated 12 patients, none of which experienced a BCAR. In response to these positive results, the UK group has just recently obtained funding from Medical Research Council of the UK to perform a follow-up trial in kidney transplant recipients; this trial is called “The TWO Study”, and will treat around 40 patients. The Boston MGH group has also just received funding from the NIH to perform a new trial in transplant recipients based on an improved yield manufacturing protocol for Ag-specific Tregs. The UCSF group has also received additional funding to perform new trials using their cell product in renal transplant recipients (cells given after approximately 6 months after transplant) and in liver transplant recipients after more than one year after transplantation.
Monotherapy conversion (immunosuppression minimization)
Notably, 12 CTG patients were reduced to receiving only Tacrolimus monotherapy starting approximately 9 months after transplantation. This is an overall rate of conversion of 32%, among cell therapy recipients. There is an uneven distribution of these patients among all the CTG trials, due to the fact that monotherapy conversion was optional at 9 months in the protocol. Nonetheless, approximately one-third of the combined cell therapy patients are on a single immunosuppressant. This proportion increases when including patients that were weaned to monotherapy after the 15 month observation period. Importantly, none of the patients on monotherapy have experienced a BCAR to date.
The ONE Study serves the important purpose to show that monotherapy may be a next stepping stone for future cell therapy protocols. Until these results became apparent, reduction down to monotherapy may have been considered not advisable by nephrologists. We believe these results set an important precident for future clinical trials aiming for monotherapy using a cell therapy approach.
Immune monitoring data for the CTG trials: Analyses of the CTG trials are still ongoing, but preliminary results indicate that Treg transfer can ameliorate alterations in immune cell composition in a dose-dependent manner. However, the final immune monitoring results for the CTG trials will not be possible until all of the data from the 6 trials is monitored completely, which we expect to be done in January 2018. Then, the analyses can be performed for each of the trials to look for effects of the various cell products on the immune system in kidney transplant recipients. It is critical that all the analyses and bioinformatics be performed on a complete set of monitored data (January 2018), since this is the only way to assure the fair comparison of results from the trials. Therefore, we will report on the immune monitoring data within the publications that will result from The ONE Study in 2018 (see deliverable D8.8 report).
Mechanistic studies and new product development. Another objective was to perform an in-depth comparative analysis of the different regulatory cell types to uncover both the common and unique mechanisms of their suppressive action. Several of our groups have been working on determining the mechanism of action of our cell products, and more advanced products have been developed from this research. For instance, the MGH group now has an improved manufacturing protocol for production of their donor specific Treg that improves yields; this new product will be used in new trials. The KCL group has tested many iterations of cell conditioning to optimize their Treg cell product (see WP1). Two of our groups have been working on the development of chimeric antigen receptor donor specific Tregs (CAR Tregs). These cell products are only in preclinical testing, but show promise to be more powerful than polyclonal Tregs. Notably, we have also conducted a workshop to test the phenotype and suppressive capacity of myeloid-derived suppressive cells (e.g. tolDC, Mreg). The results from this effort are being analysed at the current time and will be published in the upcoming months. Furthermore, the Nantes group has generated three types of regulatory myeloid cells to compare their phenotype, in vitro function and evaluate their in vivo potential of immunoregulation in a model of skin transplantation. These cells share some phenotypic markers, are all able to suppress T cell proliferation and prolong skin transplant survival in mice. Therefore, this mechanistic and product development work is innovative and continues to move the field of immunosuppressive cell therapy forward.
WP1: Cell product development/licensing
We have performed rigorous preclinical testing on hematologic cell products at seven partner locations: Regensburg (Mregs, Tregs), Charité-Berlin (Tregs), King's College-London/Oxford (Tregs), Milan (Tr1 cells), Nantes (DCs), UCSF (Tregs) and MGH (Tregs). The aim has been to license these cell products according to GMP standards and EMEA/FDA guidelines. These cell products were manufactured for use in The ONE Study (CTG) trials, started in 2014.
Development of licensed Treg cell product in Regensburg
The Treg cell product that was to be used for the induction of transplantation tolerance in Regensburg consists of in vitro expanded CD45RA+ CD4+CD25+ Treg. The goal during the first years of funding was the implementation of isolation and expansion strategies for this Treg subpopulation. For this purpose, new enrichment technologies for CD25+ cells had to be tested, flow-cytometric cell sorting of CD45RA+ CD4+ CD25+ DC127low T cells had to be established, as well as new in vitro stimulation reagents for the expansion of Treg cells. These isolation and in vitro expansion technologies for human CD45RA+ Treg cells were successfully established and meanwhile validated under GMP conditions. A license to manufacture the Tregs was obtained in the past year and the cells are presently being used in patients experiencing acute graft-versus-host disease. This trial, although not part of The ONE Study, has been approved by the Paul-Ehrlich-Institute. For two reasons, the ONE Study Steering Committee has decided not to further fund the CTG with the Regensburg Treg product. First, we currently already have two polyclonal Treg (UK and Charité) where the trials have already been initiated. Therefore, it is not necessary to repeat the testing of Tregs in a third CTG trial; in fact, The ONE Study Advisory Board suggested that funding would be better used by extending one of the two other trials (UK or Charité) with additional patients. Indeed, the UK group is poised to finish their recruitment in the first half of 2016, so if funding is available, it will be possible for this group to include more patients into their CTG trial to improve the reliability of their trial results. The Advisory Board felt that this was more important than adding another CTG trial with the same type of Treg. The second reason for this decision is that the Regensburg site alone is not currently able to supply enough patients for a second CTG trial to run in parallel to the already initiated M reg CTG trial. Therefore, Task 1 has been completed to the extent it will be taken in The ONE Study.
A) The CD25-bead-based pre-enrichment technologies for CD25+ cells have been tested and a new technology has been established in collaboration with Stage Cell Therapeutics that permits reversible CD25 labelling using magnetic bead-coated Fab-streptamers that can be removed by biotin incubation after cell selection (Stemberger et al. PLoS One 2012; 7(4):e35798). Using this technology it is now possible to sequentially label the CD25 antigen with magnetic bead-streptamers followed by fluorescent CD25 antibodies. The whole CD25-enrichment procedure has been optimized with respect to yield and purity.
B) In collaboration with BD Biosciences and Scan AG, a BD Influx high-speed FACS-sorter was reconstructed and installed into a custom-made laminar air flow cabinet. The equipment was fully qualified, including design qualification, installation qualification, operation qualification and ultimately process qualification by the generation of three consecutive Treg cell products. After finalizing these validation runs, the Government of Upper Bavaria approved high-speed flow cytometry under GMP-conditions.
C) In collaboration with Miltenyi Biotec GmbH new T cell stimulation beads have been developed and optimized for the in vitro expansion of Treg cells. Treg expansion has been validated under GMP-conditions, as well as the bead-removal technology using the CliniMACS (see list of validation tasks above). The technologies and reagents meanwhile reached market entry.
D) All required quality control tests were developed, validated and are now approved by the relevant authorities.
E) A manufacturing license has been obtained from the Government of Upper Bavaria and the use of such cell products within a clinical trial for the treatment of acute graft versus host disease has been approved by the Paul-Ehrlich-Institute.
F) Due to the recommendations of The ONE Study Advisory Board and restrictions in available patients at the Regensburg site, clinical trial authorisation for a Treg CTG was not filed with the Paul-Ehrlich-Institute in 2014.
Development of licensed Treg cell product at King's College London
Clinical grade Treg lines have been initially generated in the laboratory by enriching CD4+CD25+ T cells using clinical grade magnetic beads either with manual columns or using CliniMACS. Peripheral blood mononuclear cells (PBMCs) were depleted of CD8+ T cells followed by positive selection of CD25+ T cells. The purity (% of FOXP3+ cells) after separation was on average >60% (5 between 40-60%). The group has identified suitable culture medium, doses of rapamycin and IL-2, the time of their addition in culture, the optimal ratio of cells and anti-CD3/CD28 expander beads, and the frequency of re-stimulation for the expansion of Tregs. The Tregs were re-stimulated polyclonally twice to reach optimal production of large numbers of functional and stable Tregs. Twenty-two out of twenty-five generated Treg lines have reached a number that was compatible with clinical use (approximately 1x109), with no major differences between untreated and rapamycin-treated Tregs. In addition, the purity of the final product (rapamycin-treated Tregs) was >70% in 22 out of 25 production lines, with more than 70% of suppressive activity in 21 out of 25 lines, compared to untreated Tregs with similar purity, but significantly decreased suppressive ability. More than two years later the protocol for the clinical grade expansion of Tregs was moved to the GMP facility at Guy’s Hospital. In total, seven engineering runs using blood from healthy controls or patients were performed to optimize the separation procedure and culture conditions for Treg expansion. All engineering runs were performed in the GMP cleanroom using GMP grade reagents and consumables. Whole blood was volume reduced on a Sepax closed system cell processor followed by CD8 depletion and CD25 enrichment using CliniMACS. The expansion part of the manufacturing process was successfully transferred from the research lab to the GMP lab with no major alteration. The details for the feeding regime were finalised. Specifications for the drug product release were established. Protocols, SOPs and forms for all the procedures were written. Validations of the quality control tests to determine the specification of the final product were performed. IMPD and investigators brochure were submitted to the MHRA and the approval was granted. Ethical approval was submitted and granted as well. We have received R&D approval and a manuscript describing the GMP clinical grade preparation of the Tregs in the laboratory and in the GMP facility was published (Oncotarget 7:7563, 2016). More recently and as a consequence of some Treg product that either did not reach the number requested or expanded very well initially and then the cells died after the third stimulation, we have modified the protocol in the GMP facility in a way that if the number of cells necessary to make the product are reached before the last restimulation (day 24), the Tregs are harvested and cryopreserved until needed for infusion.
Notably, Oxford and KCL have worked together on the clinical trials. During the Brussels ‘kick-off meeting’ (13-14th April 2011) it was decided that the most effective way for the two UK groups to proceed was to work together as a single ‘virtual’ group. The cell therapy product for both UK groups is manufactured at the King’s College cGMP facility. The UK group made a single CTA application to the Medicines and Healthcare products Regulatory Agency (MHRA) in June 2013. CTA authorization was given for The ONE Study UK CTG trial in September 2013.
Development of licensed Treg cell product at the Charité-Berlin
The Berlin group has finished all testing, validations and re-qualifications required for obtaining the manufacturing license of the Treg cell product with the following characteristics:
Phenotypic characterization: >70% nTreg (CD4+CD25+FoxP3+)
Firstly, natural (n)Tregs are characteristically low in cytokine expression upon both TCR dependent (e.g. CD2/CD3/CD28-activation beads) and independent (PMA and ionomycin) cell activation. Upon stimulation, the formation of the effector cytokines IL2, TNFα and IFNγ remains low to negative, and the lack of TGFβ and IL10 formation allows further discrimination of nTregs from induced Tregs (iTregs). Secondly, the capacity of nTregs to suppress conventional T cell (Tconv) activation and proliferation can be assessed by mixed lymphocyte reactions. While the assessment of Tconv (CD4+CD25- and CD8+) proliferation in the presence or absence of nTregs is prone to errors and underlies a high intra- and inter-assay variability, CD69 and CD154 have recently been described as useful surrogate markers to assess nTreg suppressive capacities. The Berlin group has confirmed CD154 and CD69 as surface activation markers up-regulated in response to TCR activation and found in-line with previously published data that this process can be suppressed by nTregs. To functionally characterize their nTreg end-product and assess the suppressive capacity of these cells, mixed lymphocyte reactions of polyclonally in vitro expanded nTregs and freshly isolated PBMCs were performed. Co-cultures were stimulated with lethally radiated allogeneic cells over 7 hours. Subsequently, cells were stained and analysed by flow cytometry for expression of CD154 and CD69 on both CD4+ and CD8+ effector T cells. It was shown that polyclonally-expanded nTregs effectively suppress the activation of CD4+ and CD8+ Tconv, as determined by surface fluorescence intensity of CD154 and CD69. Also, it was determined that a broad TCR repertoire is preserved (heterogenous clonotype, no oligoclonality) after expansion using their protocol. The concerns of the Paul-Ehlich-Institute regulatory authority in Germany had a great influence on validation efforts for the manufacturing process. Thus, validation proceeded as follows:
Validation of bead detection and depletion: Flow cytometric methods were used in a first approach to set up a method that would allow precise detection of beads within the cell suspension. Beads in different concentrations were measured with the Navios (Beckman Coulter) and the MACSQuant (Miltenyi) flow cytometer to determine detection limits, recovery rates, linearity and precision/accuracy. Due to a high variance, the method was found not to be applicable for validation purposes. As an alternative method, the CASY cell counter was tested. Particles can be measured using an electrical current exclusion method. This method reaches high accuracy and reproducibility. However, this was not true for the beads. Bead detection was performed with a high content screener Opera (Perkin Elmer) that combines fluorescence microscopy and imaging software. Because the results obtained by this method were accurate and highly reproducible, the detection of beads by high content screening was chosen as the method of choice and validated successfully. With a robust method at hand, validation of the bead depletion was carried out and gave evidence that beads could be acceptably depleted.
Requalification of cleanroom and process equipment: To ensure compliance of the GMP site and validity of the data collected during validation runs, the cleanroom and process equipment underwent a complete requalification.
Validation of safety release criteria and aseptic handling: Release of a cell therapeutic product requires a standard release test that covers safety aspects. These include tests for endotoxin, mycoplasma and microbiological control. To ensure that the tests that are applied work in a valid manner with the test matrix, all tests were validated. Aseptic handling was validated successfully in three runs.
Flow cytometry panel: In parallel a flow cytometry panel was established that is used to determine purity and cytokine levels. The panel was set up on the basis of a flow cytometry panel that is used for monitoring patients. For purity the product is checked for CD4+/CD25+/FoxP3+ cells. The cytokine level is determined for safety reasons. IL-2, IFN-gamma and TNF-alpha were chosen to track contaminating T-effector cells. The panel was validated with respect to precision/accuracy, inter- and intra-assay variance and inter-operator variance.
Process validation: With the prerequisites fulfilled, the process validation was performed. Four consecutive runs were carried out to show that the Berlin GMP unit can manufacture a product that meets the defined criteria. All 4 runs met all acceptance criteria. The product was microbiologically clean, endotoxin free (below detection limit) and free of mycoplasma. The purity was >90%. Cytokine levels were lower than 2 percent for IL-2 and IFN-gamma, and <10 % for TNF-alpha.
Manufacturing authorisation: With these results, the Berlin GMP unit applied for the manufacturing authorisation for nTreg cells. The manufacture license for nTregs was obtained and the use of the product within The ONE Study clinical trials was approved.
Refinement of licensing for Tr1 cells in Milan
During the first year of the project the Milan group analysed the peripheral immune system of patients undergoing hemodialysis (Hemo) and reported that the system is similar to that of healthy subjects/kidney donors (Healthy) except for the higher frequency of activated dendritic cells (DC). Moreover, the ability to generate in vitro type 1 regulatory cells (Tr1) from Hemo patients is similar to that of healthy subjects.
During the second year of the project, the group further characterised the immune system of Hemo patients and reported that these patients have an increased frequency of CD11chighCD14- activated dendritic cells, being CD86+HLADR+. In addition, an increase in peripheral CD4+IFNγ+ T cells was found in Hemo patients as compared to that in healthy donors. The difference did not reach statistical significance but was clearly of interest. Additional experiments on Hemo patients confirmed previous findings regarding the ability to generate anergic Tr1 cells in vitro. However, the levels of anergy [defined as reduced cell proliferation of donor-specific Tr1 cells -above 60% - compared to donor-specific effector T cells (namely T effector cells)] was lower when PBMC from Hemo patients were challenged with healthy DC-10, as compared to PBMCs from healthy subjects. Given that: (i) Hemo patients have higher frequency of peripheral activated DC and of IFNγ-producing CD4+ T cells, which might both interfere with the in vitro generation of anergic Tr1 cells, and (ii) the population of Tr1 cells obtained at the end of the 10-day culture with donor DC-10 is contaminated also with non-CD4+ T cells (i.e. around 10% of CD8+ T cells), the group set up a new protocol for Tr1-cell generation in vitro. Namely, purified CD4+ T cells -instead of total PBMC- were challenged with donor-derived tolerogenic IL10-producing dendritic cells (DC-10). The anergized donor specific patient-derived CD4+ T cells contain an average of 7% bone fide Tr1 cells, and are termed T10 cells. The pre-Good Manufacturing Product (GMP) protocol in flasks was set up with PBMC isolated from healthy donors.
During the 3rd year of the project the group generated T10 cells from Hemo patients and determined that the release criterion to define T10 cells suitable for in vivo injection is anergy ≥60%. The protocol was then successfully reproduced by the scientists at the GMP facility in pre-GMP conditions in two separate experiments, thus demonstrating the possibility to transfer the protocol from the lab to the GMP facility. The patients enrolled in the ONErgt11 trial (RGT) in 2013 were also enrolled in a parallel study (named OSR-THE ONE). This protocol allowed the collection of one leukapheresis (LA) from both the Hemo patients receiving a kidney graft from a family related living donor and the donor. The LA product was delivered to the lab for testing T10 cell generation following the protocol developed during year 2. Four LA products were received (2 from Hemo patients and 2 from kidney donors). T10 cells could not be generated from the first couple (R1001 – HLA haploidentical), both from PBMC and from LA, likely due to limited HLA mismatched between patient and donor. T10 cells were successfully generated from the second couple (R1002) from both PBMC and LA. Thus, the group decided in the future to include in the CTG trial, an eligibility test (i.e. generation of T10 cells with PBMC) prior to collecting LA for GMP preparation.
Based on the successes achieved during the 3rd year of the project, the group in the 4th year collected all relevant information and submitted pre-submission documents to the local regulatory authority (Istituto Superiore della Sanità, ISS) and requested a pre-inquiry meeting. The pre-inquiry meeting was positive and they were invited to perform 3 validation runs to compile the final dossier to be submitted to ISS for final approval. Thus, in January 2014 the first validation run was planned. The GMP protocol for the generation of donor-specific T10 cells envisaged the use of fetal bovine serum (FBS). Unfortunately, the GMP-grade FBS that was chosen for their protocol was out of market and for the following 3 months they attempted to get GMP-grade FBS from alternative sources with no success. Thus, the group took the hard decision to change from FBS to Human Serum (HS) and, therefore, had to perform additional experiments to test the new protocol in the lab. T10 cells were efficiently generated by using buffy coats and HS (n=3 experiments – April-June 2014). Accordingly, they decided to change the GMP protocol for the generation of DC-10 cells from FBS to HS. The first validation run was set for July 2014. Donor specific anergic T10 cells were obtained. However, the DC-10 yield was extremely low and monocytes from leukapheresis had an activated phenotype. The GMP facility then had to close down for renovations (August-December 2014).
On January 2015 the lab, based on the data generated using buffy coats, defined the FBS as key component for the DC-10 generation from monocytes. Thus, a second validation run with FBS was set and performed but still the DC-10 yield was extremely low and incompatible with a future cell therapy trial. At that stage, the group went back once more to the lab to define the critical steps in DC-10 generation in a GMP facility (i.e. cliniMACS vs lab-based monocyte purification, leukapheresis vs buffy coats being the most relevant differences). A set of experiments performed in the lab together with a continuous collaboration with the GMP facility led to the conclusions that the monocytes purified from leukapheresis by cliniMACS and cultured in standard cell culture flasks in the GMP facility had a completely different phenotype and in vitro behaviour than those isolated from buffy coats and purified with lab-grade systems. Promising results were obtained by changing the culture support: the group decided to move from flasks to DC-culture bags (from Miltenyi GMP grade). On June 2015 the 3rd validation run was performed with great success and on October 2015 the 4th validation run was performed confirming the definition of a feasible and effective GMP-grade protocol for the generation of Ag-specific T10 cells from patients with kidney failure.
In conjunction with this preparatory work, the Milan group has:
1) demonstrated that the GMP protocol previously set up for the ALT-TEN trial performed in haematological cancer patients to prevent GvHD (Bacchetta et al Front in Immunology 2013) is not suitable for the generation of donor specific T10 cells using blood of patients with renal failure. All the subsequent changes for the development of a suitable protocol specifically tailored for patients enrolled in the ONE Study have been collected and published (Transplantation 99:1582, 2015).
2) gathered all the steps necessary to translate a protocol from a GMP (the one used in the ALT TEN trial) back to the lab and tailored for specific patient characteristics and then back again to a new GMP facility (Mfarrej B et al. manuscript in preparation).
This preparatory work is contributing to a submission to the authorities for the new cell manufacturing procedures. A clinical trial has therefore not been initiated in Milan.
Development of licensed M reg cell product in Regensburg.
All essential steps in the development of Mreg_UKR (which is the name of the Mreg cell product) were completed in 2013. Manufacturing licenses for production of monocyte apheresates and generation of Mreg_UKR from monocytes were granted by the Regierung von Oberbayern on the 10th September and 8th August, 2013, respectively. All quality, preclinical and clinical development data have now been approved by the German competent authority (PEI), in support of the ONEmreg12 CTG trial. First versions of the Investigation Medicinal Product Dossiers (IMPDs) and investigators’ brochure were submitted to PEI on 5th June, 2013. The PEI responded on 28th August, 2013 with a list of deficiencies, which were addressed in revised versions of the IMPDs and IB, submitted on 22nd November, 2013. The PEI gave full approval for the trial in December 2013.
This decision was forwarded to the local ethics committee in Regensburg, which demanded substantial amendments. The required changes were approved by the local ethics committee and forwarded to the PEI. A positive response to the amendments was granted by PEI at the end of April 2014 and the ONEmreg12 trial was formally initiated thereafter. The eCRF was ready for operation in March 2014, and patient recruitment could then begin. The first study patient was recruited in July 2014 but the cell product did not pass the release criteria and could not be infused. A substantial amendment relating to release specifications was granted by PEI in October 2014. A second patient was recruited to the study and treated with Mreg_UKR later in the same month. This was followed by another release failure in January 2015 (3rd patient) and a cancellation of the manufacturing run for the 4th patient in February owing to release assay problems identified at the manufacturing facility. To ameliorate these release assays problems, two substantial amendments were submitted to the PEI (in March and June 2015). The second of these proposals was approved in July 2015, after which the subcontractor Apceth validated the new release assay methodologies. A 5th patient was recruited in October 2015 but withdrawn immediately due to a violation of the trial eligibility criteria. The 6th patient could not be treated because the cell product failed release testing. A 7th patient was enrolled and treated with Mreg_UKR in April 2016. An 8th patient was enrolled in February 2016 but could not be treated owing to another release failure.
Clinical development of Mreg_UKR is an ongoing process. During 2016, non-clinical investigations were performed in an attempt to determine the underlying cause of the cell product manufacturing failures. It was found that all 3 charges of human AB serum used during cell culture (during process development and clinical production runs) contained detectable levels of anti-HLA antibodies. The presence of these antibodies did not appear to disturb M reg development in culture, as demonstrated by several successful manufacturing runs when the final product passed the release criteria, however it cannot be completely excluded that these Abs might have an adverse effect on M reg survival/function following administration of Mreg_UKR to patients. The trial will continue to use human AB serum for future Mreg_UKR manufacturing runs. New charges of serum will be screened for anti-HLA antibodies and cross-matched against donors. This means that the study will proceed using cross-match negative serum, as opposed to bead array-negative serum. An informational letter on this topic is being published in the journal Transplantation (currently in press).
It is important to add to this report that much work was required in response to PEI’s reply to our first submission in 2013. It was necessary to perform additional work in all areas of the application; in particular, comments concerning the preclinical pharmacological and toxicological development of Mreg_UKR necessitated that new experiments were performed. With regard to quality and manufacturing, the PEI was especially concerned that intermediate- and release-specifications should reflect the process of differentiation of Mreg_UKR from monocytes; therefore, new specifications based on CD14 down-regulation were developed and validated. The PEI requested a more extensive routine characterisation of cellular contaminants of Mreg_UKR products, notably of CD62L+ thrombocytes, which the PEI claimed may be thrombogenic; therefore, new specifications were developed and validated. As mentioned above, the PEI demanded further results from the validation of microbiological controls of intermediates and the final product. With regard to preclinical development, the PEI asked questions concerning primary pharmacodynamics, secondary pharmacodynamics, safety pharmacology and toxicology. Questions about the primary pharmacodynamic effect of Mreg_UKR were answered using unpublished data showing that Mregs drive CD4+ T cells to become FoxP3-expressing regulatory T cells; these data have now been incorporated into a scientific manuscript that is being expanded for submission. Additionally, a systematic and unbiased comparison of human and mouse M regs was made in order to justify the conclusion that therapeutic effectiveness in mice predicts effectiveness of Mreg_UKR in humans; specifically, they compared iNOS and IDO expression in mouse and human M regs, as well as other markers by flow cytometry and whole-genome gene expression profiling. Questions regarding safety pharmacology were answered by performing new studies according to ICH Topic S 7 A Safety Pharmacology Studies for Human Pharmaceuticals (CPMP/ICH/539/00). Questions from the PEI relating to the toxicological and virological safety of monocytes sorted with CD14-microbeads were answered with data provided by Miltenyi GmbH and experimental data generated in collaboration with Professor Barry Sharp (LU) through WP2. In response to issues raised by the PEI concerning the safe application of Mreg_UKR, extensive changes were implemented in the clinical trial protocol and clinical IMPD. In particular, clinical strategies were developed for: management of possible microbial contamination of Mreg_UKR products detected after administration; identification and treatment of non-responders to Mreg_UKR; emergency and non-emergency management of adverse reactions against Mreg_UKR; and the definition of ‘critical pathways’ in the clinical delivery of Mreg_UKR treatment at our hospital. Some of these aspects have been included in a manuscript that was recently published in the journal Kidney International (Hutchinson JA and Geissler EK, June 2015) and one which was published earlier in September 2014 in the journal Transfusion (Vol 54: 2336-2343).
Development of licensed DC cell product in Nantes
During the first year, the group defined the protocol to derive TolDC from human blood. From leukapheresis product of peripheral blood, monocytes are enriched by elutriation (purity around 90-95%) and cultured for 6 days in AIMV medium supplemented with GM-CSF. Analysis of phenotype and function of these cells were assessed: TolDC are HLA-DRlowCD80lowCD86lowCD14+CD209- cells. About their function, TolDC induce a weak stimulation of allogeneic T cells compared to conventional DC and are resistant to maturation induced by LPS+IFNγ stimuli. TolDC are also able to inhibit proliferation of T cells in response to an allo-stimulation (allogeneic mature dendritic cells); this inhibition assay is described in WP5.
During the second year, the group translated this work into the clinical protocol of TolDC generation. Patient cells will be produced in the GMP facility called UTCG (Cell and Gene Therapy Unit, Director: Prof. Brigitte Dreno, http://www.ifr26.nantes.inserm.fr/utcg/?lang=en) located in Nantes Hospital. In close collaboration with UTCG, the manufacturing process was studied and comparative studies of containers (plates/bags/flasks) and media (AIMV/RPMI albumin) were performed to determine the optimal culture conditions. These studies showed that culture in 6-well plates and the use of AIM V medium allow the generation of TolDC in compliance with our phenotypic and functional criteria. Stability of TolDC was also tested over 6 hours. At the end of the second year, the manufacturing process of TolDC according to GMP condition was defined, as well as quality and microbiological controls.
During the third year, the group validated the generation of cells from patients suffering from kidney insufficiency (non-interventional study) and performed validation runs from blood samples of healthy donors and from one patient (1 body mass) to validate the generation of TolDCs by the UTCG.
During the 4th year, the last validation run was performed and the application forms (IMPD and clinical protocol) for The ONE Study ATDC trial were submitted at the beginning of February 2014 to the competent authority ANSM (National Agency for Medicinal and Health Product Safety). ANSM responded in June 2014 with a list of deficiencies, which were addressed in a revision and submitted in July 2014. All approvals by the ANSM were received in September 2014. The ethics committee also approved the CTG trial in July 2014.
During the end of the funding period, they continued to investigate the mechanisms of T cell suppression mediated by ATDCs. They showed that ATDCs are able to decrease Th1 and Th17 differentiation and promote Treg expansion. Furthermore, they identified that ATDC mechanisms of action was not dependent of cell contact. The soluble molecules involved in ATDC mechanisms are still under investigation.
Development of licensed Treg cell product with donor specificity at UCSF
At UCSF, significant progress has been made in developing a GMP compliant, FDA approved product for the cell therapy phase of The One Study. They have established a methodology to expand Treg (100 to 4,000-fold) from peripheral blood or leukapheresis products. The manufacturing process begins with flow cytometer purified Tregs based on the phenotype of CD4+, CD25hi, and CD127lo expression. These sorted cells are initially stimulated with donor B cells that have been activated with GMP-compliant K562 cells expressing human CD40L, to selectively stimulate donor alloantigen-reactive (dar) Tregs. The cells then are re-stimulated with CD3/CD28 beads along with IL-2. The cells are highly functional in in vitro suppression assays and have a high degree of alloreactivity. Greater than 75% of cells express FOXP3 and demonstrate de-methylation of the foxp3 locus by the TSDR assay, perhaps the most reliable marker for stable Treg. This manufacturing process has recently been published in the American Journal of Transplantation (PMID 24102808). This process has been submitted as a drug master file (DMF) to the FDA in the context of an Investigational New Drug (IND) application for a planned phase 1 trial in liver transplantation. An IND application for the use of these cells in living donor kidney transplantation as part of The ONE Study consortium was submitted to FDA in June 2014 and the FDA informed the team in July 2014 that the IND (16043) was approved. Therefore, aspects of this task have been completed and the trial has been initiated.
In the course of the cell therapy group enrolment at UCSF, the cell manufacturing group came to the realization that the rate of EBV DNA positivity can be an impediment to trial progression. Despite the use of gangciclovir in the cultures, 3 out of total of 11 clinical-grade B cell cultures had detectable EBV DNA in the culture supernatant. It was not clear whether the DNA was from infectious viruses reactivated in culture, non-infectious virion-free EBV genome release by dead B cells, or was explained by false positive testing. Extensive further studies in 2017 finally revealed that the testing was actually a false positive result and it was determined that the previous preparations could have been used.
Development of licensed Treg product with donor specificity at MGH.
Previously, this research group has conducted three phase I clinical trials in hematopoietic stem cell transplantation for hematologic malignancies in which donor bone marrow was cultured for 36 hours with recipient antigen-presenting cells in the presence of CTLA4Ig (co-stimulatory blockade) prior to infusion (1 study) or donor marrow or peripheral blood mononuclear cells (PBMCs) were cultured for 36 hours with recipient antigen-presenting cells in the presence of monoclonal antibodies to CD80 and CD86 (co-stimulatory blockade) prior to infusion. These cell products were shown to be safe, and graft-versus-host-disease incidence was less than historic controls. In subsequent studies, it was determined that this approach results in an increase in donor (ie responder) Foxp3+ T cells both in vitro by the end of the culture and in the cell infusion recipients, and that these cells exhibited antigen-specific suppression. In The ONE Study, the MGH group is adapting this approach to solid organ transplantation, to develop recipient T regs specific for donor antigens. As the starting population is unfractionated cells, the Tregs may include both expanded natural Tregs as well as induced Tregs, representing a diverse repertoire. In the past year, the group has used the FDA approved drug Belatacept for costimulatory blockade in cultures of unfractioned T cells stimulated with irradiated allogeneic PBMCs. Importantly, they have conducted cultures using T cells from uremic patients on dialysis, to adequately model the clinical situation which will be encountered in the trial. In their protocol, cells are cultured for 72 hours and then after extensive washing passed over successive Miltenyi CliniMACS columns for isolation of CD4+ cells, followed by isolation of CD25+ cells (from the CD4+ fraction). The recovered cells are ~96% CD4+ and at ~90% CD25+CD127-Foxp3+ are characteristic of Tregs. These cells have also been analysed by Epiontis for demethylation at the FoxP3 TSDR, a feature which is believed to be the hallmark of bona fide stable Tregs. They find an average of 97.1% of the cells are demethylated. Standard operating procedures for cell production, the clinical study protocol and both donor and recipient consent forms have been written in support of a Phase I patient safety study at a Treg dose of 104/kg-2x106/kg. This group has received local ethics approval, as well as an approved IND from the FDA. They have enrolled three patients to date. Two of these patients have been minimized successfully to Tacrolimus monotherapy (MMF has been weaned away). However, the MGH group decided that the cell yields were within acceptable limits for infusion, but were below their expectations. A decision was made to stop their ONE Study trial, and redesign the manufacturing procedures to improve yields. These procedures have now been improved and will be used in a new trial that has just received funding in the USA.
Supportive work by Miltenyi.
In the first period Miltenyi has worked on the development of GMP processes for isolation of Treg in sufficient quantity and quality to allow further in vitro expansion. They have improved the CD25 sorting protocol via CliniMACS to achieve purities of Foxp3+CD127- Treg >70-80% (previously, 50% on average). In close cooperation with the partners in Oxford, London, Berlin and Regensburg, they developed a general strategy for Treg sorting which is now part of the clinical protocols used in the three centres Oxford, London and Berlin. This strategy includes CD8 depletion prior to CD25 enrichment. Treg separation has been validated in at least 20-30 large scale experiments using leucapheresis material and the CliniMACS instrumentation, providing a solid data basis for further validation of the production protocols in the various local centres.
In the second period, Miltenyi continued the development of GMP Treg expansion reagents suitable for expanding Treg with high yield and purity. These CD3 and CD28 -coated particles have been tested by several consortium members and will be used for the clinical expansion protocols. They are now available as a GMP-compliant reagent. In addition a clinical scale bead removal procedure has been developed, which allows the efficient removal of CD3/CD28 particles before injection of Treg into patients. This has proved to be a key step required by the regulatory authorities.
A regulatory support file has also been generated by Miltenyi, documenting all safety and quality control issues of the Treg expansion system; this file has been made available by Miltenyi to the regulatory authorities to facilitate approval of the planned cell generation protocols.
In addition, Miltenyi has developed a specialized GMP T cell expansion medium and tested additives such as rapamycin. Their results indicate that the use of rapamycin allows the reproducible generation of high purities of expanded Treg (70-90% FOXP3+). Miltenyi is also working on the development of a large scale Treg expansion protocol testing various culture vessels and culture conditions. These tests are still ongoing, but a basic protocol is already available that can be used for setting up cultures for clinical trials. All expansion protocols have been done with CliniMACS isolated Treg from leucapheresis products, i.e. using exactly the same material that will later be used for the clinical trials. So, again, this is a solid data basis for further validation experiments required for licensing the expansion protocols at the various centres.
In the third year Miltenyi further focussed on improvements for the expansion of human Treg, including improved sorting strategies, automated systems, improved cell culture tools and reagents. A second focus was the development of strategies for antigen-specific Treg (see WP6). A major issue towards broad application of cellular therapies is the availability of automated systems for cell generation. Miltenyi has released a first fully automated clinical cell sorting device, the CliniMACS Prodigy, which allows automated density centrifugation, labelling, separation and subsequent expansion of separated cells. Currently, Treg separation protocols for this automated device are developed, following the strategies developed during The ONE Study, i.e. CD8 depletion plus CD25 enrichment. In addition, cultivation protocols for Treg are developed in this automated device. This includes testing of various components which shall facilitate automation for vitro culture, i.e. new cell culture devices containing membrane bottoms for improved gas exchange or new CD3/CD28 activation systems based on nano-sized carrier systems which simplify dosing and bead removal in comparison to the currently used micron-sized particles. Miltenyi has also tested whether pre-selection of Treg subpopulations via magnetic cell separation may improve the expansion results. However, CD45RO depletion before CD25 enrichment did not result in improved expansion or purity. Therefore, these approaches are not planned for further development.
In the fourth and fifth period Miltenyi proceeded with the development of applications on the new GMP compatible device for flow cell sorting. The "MACSQuant Tyto" is designed for clinical flow-sorting in a standard GMP facility without the need for specialized safety cabinet systems. With this device clinical sorting of CD25+CD127- Treg +/- CD45RA can be envisaged which will facilitate clinical cell sorting and expansion under routine conditions. Miltenyi has developed reagents for GMP-compatible flow-sorting of CD25+ CD127- CD45RA+/- Treg. Prototypes of the “Tyto” are currently tested by members of the consortium and the device are now available and being used by the KCL site. Miltenyi is also evaluating the combined use of CD154 and CD137 for clinical flow-sorting of stable Foxp3+ Treg from mixed cultures, which can improve the purity and stability of the Treg infusate (see also WP6).
In the final part of The ONE Study, Miltenyi has worked on the MACSQuant Tyto sorting process using the GMP compatible reagents allowing to sort CD127-CD25+CD45RA+/- Treg by the new GMP sorting device. The current sort strategy includes CD25 enrichment via CliniMACS and subsequent flow-sorting. First test runs have successfully performed and currently the process is optimized including the subsequent expansion of the sorted Treg populations. For the in vitro expansion of Treg Miltenyi is also testing a new nanosized activation matrix (TransAct reagent), overcoming the difficulties associated with micron-sized particles which are currently used for expansion of T cells for clinical application. The TransAct reagent is ideal for use in an automated and closed-system cell culture device such as the CliniMACS Prodigy. The TransAct reagent is already being used for clinical scale expansion and viral transduction of conventional T cells. The protocol is currently being adapted for human Treg and the tests are ongoing. Miltenyi has also evaluated the combined use of CD154 and CD137 for enrichment of stable Foxp3+ Treg from mixed Treg cultures. The results clearly show that CD137 expression and lack of CD154 expression identifies a Treg with a highly demethylated TSDR region, so far the best molecular proof for stable Treg. This sort strategy will allow to enrich for stable Treg even following extended Treg culture, e.g. in the course of viral transduction with artificial antigen receptors (see also WP6). Overall the improved sorting strategies and automated cell culture of Treg for generation of clinical Treg products will greatly facilitate eventual performance Treg cell therapy as a standard clinical procedure.
WP2: In vivo tracking of administered cells
Determining the fate of therapeutic cells after administration to patients is central to understanding their potential immunomodulatory effects in vivo. In the context of regulatory cell populations administered by intravenous infusion, subsequently detecting the cells presents a number of technical challenges. Firstly, the cells of interest are massively infrequent compared to the recipient’s own leucocytes or tissue cells. Secondly, to discriminate recipient-derived therapeutic cells from the recipients’ own cells, or to discriminate donor-derived therapeutic cells from graft-derived cells, the infused cells must be marked with a label which should not be readily transferred between cells and should not affect the cells’ regulatory functions. The objective of work package 2 (WP2) was to develop a sensitive method for tracking cells after infusion into patients, based on laser ablation - inductively coupled plasma – mass spectrometry (LA-ICP-MS) detection of labels.
A series of experiments were conducted to optimise parameters for labelling T cells with Gd-chelates and labelling Mregs and Tol-DC with gold nanoparticles. Cells were labelled in Oxford, Regensburg and Nantes under various conditions and shipped to Loughborough University (LU) for analysis. Population analysis revealed an uptake of approximately 108 atoms of Gd per cell and up to 109 atoms of Au per cell. These high cell loadings enabled single cell detection of labelled cells using LA-ICP-MS. Furthermore, the label remained present and detectable for a period of at least ten days after labelling. An alternative metal-bound antibody labelling approach was used to enable the detection of CD45, CD11c and HLA-DR in single Mregs. It is anticipated that this approach could be used to post-label samples in order to establish that cells bearing the tracer label exhibit the anticipated marker phenotype.
Flow cytometry analysis indicated that labelling cells with 50 nm gold nanoparticles does not affect their expression of CD11b, CD14, CD16 or HLA-DR. Furthermore, gold labelling did not affect the capacity of Mregs to suppress polyclonal proliferation of PHA-stimulated allogeneic CD4+ cells in direct 1:1 co-cultures. No difference in viability was observed between labelled and non-labelled Mregs. Similarly, Gd labelling did not affect the viability of T cells in vitro. Studies in CBA Rag KO – (H-2k) mice, carried out by the Oxford group, indicated that Gd-labelling does not affect the ability of nTreg to regulate rejection of a B6 (H2-b) skin graft. Therefore, at present there is no evidence that either of the labelling approaches interfere with regulatory cell function.
The labelling approaches outlined above were used to perform short-term tracking studies in animal models (up to 10 days). Labelled T cells were administered to immunodeficient mice and subsequently analysed on a single cell basis in ex-vivo samples. The cells were successfully identified in peritoneal lavage samples at 3, 6 and 10 days post-administration. Labelled Mregs were also successfully detected in ex vivo mouse samples at 24 hours and 7 days post-administration, with the lungs, liver and spleen identified as the major sites of M reg accumulation. This was in agreement with separate flow cytometry experiments performed by the Regensburg group. Subsequent laser ablation analysis of thin sections of liver, lung and spleen tissue pinpointed the spatial location of the Au labelled cells.
The above experiments demonstrated that the minimum requirements for single cell detection could be met using conventional LA-ICP-MS technology. However, the throughput of this existing technology was limited by the gas flow dynamics of laser ablation chambers. A prototype laser ablation chamber, designed and tested at LU, provided a 50-fold increase in speed and a six-fold increase in sensitivity for the analysis of Gd labelled T cells. ESI have now integrated the new chamber into their systems to produce a robust and user-friendly product, the NWR image.
Development of the NWR image focussed on four key areas: spot size/resolution, laser wavelength/pulse rate, high-resolution sample viewing and, as mentioned above, sample transport speed. The resulting laser operates at 266nm (frequency quadrupled from a Nd:YAG diode-pumped source) to achieve the best coupling with organic material, while maintaining integrity of the glass substrate. A novel objective lens array was designed and implemented into the platform to achieve both high-quality delivery of the UV beam to the sample surface and high resolution optical viewing of the sample. The laser beam delivery system can achieve a spot size range of sub-micron to 50 micron per ablation. This was challenging because it approached the diffraction limit for light passing through an aperture, but was an essential requirement to adequately distinguish neighbouring cells in tissue sections.
The data from the cell labelling work has been published in four papers (Managh et al., Analytical Chemistry, 2013, 85:10627-10634, Managh et al. Journal of Immunology, 2014, 193:2600-2608, Hutchinson et al. Transplantation Direct, 2015, 1:e32 and Hutchinson et al. Transplantation, 2015, 99:2237-8). A fifth publication covered improvements to the underpinning laser ablation technology (Douglas et al. Analytical Chemistry, 2015, 87:11285–11294).
WP2 has had a significant training impact on a researcher who was recruited as a PhD student and has since gone on to secure a permanent academic position at LU. Further, the technology developed during WP2 has since been commercialized by ESI and is already being adopted for a variety of bio-imaging applications. This is a major accomplishment of this work package. The technology is indeed very promising for the original purpose of detecting labelled cells after cell therapy and we remain confident that this has a high potential impact for eventually tracking cells in humans.
WP3: Trial development, performance and management
The ONE Study involved the design, approval and performance of 7 individual clinical trials.
Trial 1. The Reference Group Trial (RGT) was planned during the first year of The ONE Study funding period. This trial serves as a control or reference group for the 6 Cell Therapy Group (CTG) trials that were not planned to start before the 4th funding year. The RG T design was agreed upon by all 8 transplant centers involved in The ONE Study, and received ethical and regulatory approval by all involved authorities between September 2012 and September 2013. Patient recruitment was completed in September 2014, with the aim of recruiting 60 informative patients. In total, 70 patients were enrolled into the RGT, with 9 patients dropping out prematurely; therefore, 61 informative patients were transplanted and followed-up until the final trial visit (i.e. 60 weeks post-transplantation). The final RGT study visit (Last Patient Last Visit) took place in December 2015.
All clinical data has been monitored/verified and secured in the eKoehler data capturing eCRF data base, as planned.
An extensive effort was made to create a portfolio of standardized & reproducible molecular assessments, multi-parametric flow cytometry as well functional assays which are now used by many laboratories world-wide. Analysis of the immune monitoring data has resulted thus far in the following major general findings (final data analyses are planned to be complete in January/February 2018):
RGT: Combining the findings of all RGT patients we identified a set of biomarkers (e.g. Nav3 gene expression), which allows pre-transplant identification of patients experiencing acute rejection. In addition, we detected alterations in peripheral blood immune cell composition (e.g. % of total B cells, naïve CD4+ T cells) in patients showing a decline in graft function up to 12 months prior to clinical manifestation. The biomarkers are expected to be useful in the future for the detection of immune tolerance or allograft rejection. Although the RGT was meant only as a comparator for the CTG trials, we now realize that there is a wealth of information from the extensive and stringent immune monitoring that we performed. Such an analysis of an international cohort of normal kidney transplant recipients has not been done before. We expect these results to serve also as a reference for future studies that use a similar protocol design (e.g. similar trial design now being adopted and used by Prof. Angus Thomson, University of Pittsburgh, testing donor-derived tolerogenic DC).
Trials 2-7. The CTG trials
There are 6 CTG trials that were initiated within The ONE Study. Two trials tested the use of polyclonal T regulatory (Treg) cells (either frozen or fresh cells), two trials use donor-reactive T regulatory cells produced by two different methods, and two trials use blood monocytes as starting material to produce either tolerogenic dendritic cells (tolDC) or regulatory macrophages (Mreg).
All 6 CTG trials were initiated sometime within approximately one year after April 2014. All trials followed the same clinical trial protocol (tacrolimus+MMF maintenance, +3 months of steroid treatment), with the exception of delivery of a different cell therapy product. Also, different cell numbers were delivered in each trial and the timing of delivery relative to kidney transplantation was specific for each trial. A 7th trial was planned by our partner in Milan, where they would use a cell product enriched for Tr1 cells (another donor-reactive T regulatory cell product), but this trial was not initiated because of issues related to obtaining the approval for the trial/cell product manufacturing; this trial has not yet started and will not be part of The ONE Study group of cell therapy trials.
Planned patient numbers to receive cell therapy for each cell product varied between 6 and 12. At the writing of this report, 5 of the 6 trials have either reached completion or have been stopped. Here is a brief summary of each of the trials and their current status:
• UK trial (Oxford and London): Polyclonal Treg-frozen product
The planned 12 patients were treated with polyclonal Tregs, and all patients have completed the 15 month follow-up phase of the trial. In total, 15 patients were recruited into the trial, with 12 actually receiving Treg; 3 cell manufacturing failures occurred. The trial has reached completion and all data in the eCRF has been monitored. Two patients were reduced to Tacrolimus monotherapy after 9 months posttransplantation. Notably, reduction to Tacrolimus monotherapy at 9 months posttransplant was optional within our study protocol.
• Berlin trial: Polyclonal Treg-fresh product
11 patients were treated with polyclonal Tregs, according to plan, and in December 2017 all patients have completed the 15 month follow-up phase of the trial. In total, 17 patients were recruited into the trial, with only 11 actually receiving Treg due to some manufacturing failures. The trial has reached completion, but some data in the eCRF still needs to be monitored, which is expected to be complete in January 2018. A high proportion of patients, 8, were reduced to Tacrolimus monotherapy.
• UCSF (San Francisco) trial: Ag (donor)-specific Treg product
2 patients were treated with the UCSF donor-reactive Tregs, and both patients have completed the 15 month follow-up phase of the trial in December 2017; eCRF data monitoring will be completed in January 2018. In total, 5 patients were recruited into the trial, with 2 actually receiving Treg; 3 cell manufacturing failures occurred. It should also be noted that a false signal of EBV contamination was the reason for products failing release; after a long investigation, it was determined that this was a false positive signal. This investigation did create delays in recruitment of more patients. Notably, one of the two patients (the second patient infused) that was treated with cells did show a substantial infiltration with lymphocytes upon kidney biospy shortly after Treg infusion, which was is arguably evidence of BCAR; however, the central pathologist has not yet confirmed the diagnosis of BCAR. Nonetheless, there was enough concern by the local investigators that infusion of donor-specific Treg under this protocol should not be continued, even though the data safety monitoring board did not suggest to discontinue the study. Therefore, the decision to stop recruitment of more patients was made by the local investigators. No patients were reduced to Tacrolimus monotherapy. As a result of The ONE Study experience with this cell product, future trials by the UCSF group using these cells will not likely utilize cell infusion near the time of organ transplantation.
• MGH (Boston) trial: Ag (donor)-specific Treg product
Three patients were treated with Ag-reactive Tregs, and all 3 patients have completed the 15 month follow-up phase of the trial, with eCRF monitoring currently underway and scheduled for completion in January 2018. In total, 5 patients were recruited into the trial, with 3 actually receiving Treg; 2 cell manufacturing failures occurred. Two of the patients were reduced to Tacrolimus monotherapy, and the third patient was also reduced to monotherapy, but after the 15 month follow-up period. The investigators elected to not recruit more patients into the trial because the yield of Ag-reactive Tregs with their existing manufacturing protocol was deemed to be too low. They decided to establish an improved manufacturing procedure that will bring a greater yield of Ag-reactive Tregs and start a new trial. Indeed, the new manufacturing procedure is ready now and a new trial has received funding.
• Nantes trial: tolDC product
Eight patients were treated with tolDC, and 5 of the 8 patients have completed the 15 month follow-up phase of the trial; 3 remain in follow-up that will extend past the end of The ONE study. The eCRF monitoring for the patients that have completed the 60 month follow-up will be finished in January 2018; the remainder of the patients will be followed-up in due course until they have reached the 60 month mark. In total, 11 patients were recruited into the trial thus far, with 8 actually receiving tolDC; 3 cell manufacturing failures occurred. None of the patients have been reduced to Tacrolimus monotherapy. The investigators currently have their trial on hold, pending additional funding for possibly treating additional patients.
• Regensburg trial: Mreg product
Two patients were treated with Mreg, and both have completed the 15 month follow-up phase of the trial. In total, 8 patients were recruited into the trial, with 2 actually receiving Mreg. The eCRF for both patients has already been monitored. There were two problems with this trial that limited the number of patients able to be treated. First, there were problems with the cell manufacturing that resulted in several of the recruited patients not being treated. These manufacturing problems were solved over the course of approximately 1 year, but then there was an issue at the site regarding patient eligibility over the past year; while over 20 patients were screened for inclusion into the study in 2017, none of them fulfilled the inclusion criteria (an explanation for this is provided in the report for deliverable D4.2). Neither of the two patients have been reduced to Tacrolimus monotherapy. The study has effectively been halted at this point, due to a funding gap, since The ONE Study funding period has ended. It is also notable that significant financial resources for engineering runs of cell products were necessary to solve the cell manufacturing problems.
Immune monitoring results for CTG trials: Analyses are still ongoing, but preliminary results indicate that Treg transfer can ameliorate alterations in immune cell composition in a dose-dependent manner. However, the final immune monitoring results for the CTG trials will not be possible until all of the data from the 6 trials is monitored/verified completely, which we expect to be completed in January 2018. Then, the analyses can be performed for each of the trials to look for effects of the various cell products on the immune system in kidney transplant recipients. It is critical that all the analyses and bioinformatics be performed with a complete set of monitored data (January 2018), since this is the only way to assure the fair and consistent comparison of results from the trials, which was a primary objective of The ONE Study.
Trial performance conclusions:
Regarding the RGT, 70 patients were recruited and this trial was completed as planned. For the CTG trials, 60 patients were recruited into the various trials, with 38 of these patients being treated with one of the 6 cell therapy products. All of these trials are either completed with recruitment, or have been stopped due to the reasons described above in the report. As a lesson from The ONE Study, excellent recruitment in the UK by the combined Oxford and London partners bodes well for sharing efforts towards recruitment. Importantly, the framework for The ONE Study worked perfectly in terms of supporting the initiation of the trials. Groups freely shared their IMPD/IND documents with each other to get approval from their authorities and all groups have realized that performing these trials would have been impossible in this time frame without the support of the whole ONE Study consortium. In group meetings, we estimate that it would have taken approximately 3 more years to start and complete trials without the cooperation of our partners. Indeed, some of the groups believe they would not have even gone to the level of clinical trial performance without The ONE Study support. Therefore, the structures and mutual support provided within this consortium has opened the way to the performance of new trials in organ transplantation that would not otherwise have been possible. “The ONE Study model” is being copied and promoted by other funding organizations (e.g. Immune Tolerance Network) to organize new efforts in transplantation and cell therapy.
WP4: Trial evaluation
The primary objective of The ONE Study was to test the comparative safety of various T cell and monocyte derived immunoregulatory cell therapies in renal transplant recipients. The strategy for properly testing safety was to test the cell products using the same basic clinical trial protocol, so that any deleterious effects could be equally assessed. The safety assessment was expected to be one of the most critical outputs from The ONE Study, so that new trials could then be based on cell products that already have an existing safety profile.
Electronic Data Collection System:
The electronic data capture system (eCRF) has been completely developed and implemented for the RGT; the web-based system was released in August 2012 and was used by all participating trial sites.
More specifically, KOEHLER eClinical has been an essential partner to develop this novel type of database whereby all clinical and immune monitoring (e.g. microarray data, FACS) information are incorporated. KOEHLER eClinical’s market-introduced products include a web-based-platform which can be used for the conduct of clinical trials and all relevant processes. This platform has been used during The ONE Study RGT and CTG trials and is set-up for the specific protocol requirements and customized to meet the particular needs of each partner and the consortium as a whole. This system serves as a centrally accessible portal to view the safety data collected in all trials within The ONE Study. These systems were set up and available as planned for each of trials before they were initiated. The design of the eCRF is valuable, and will be offered to the public in the form of a publication that we are planning; the idea is to allow others conducting trials to use the same system.
Safety - The Reference Group Trial (RGT):
The RGT utilized a standard immunosuppressive treatment regimen in living donor kidney transplant recipients (Basiliximab induction and 3 months steroids + maintenance therapy with Tacrolimus and MMF). This regimen is already approved for use in kidney transplantation, so the RGT was performed with the normal reporting of adverse events through a pharmacovigilance system that we established for this study in Regensburg. This electronic pharmacovigilance system (VigilanceONE) was made fully operational and was tested in the RGT. However, since there were no safety concerns anticipated with using this standard protocol, no special data safety monitoring board was established. We simply reported the adverse events (AEs), serious adverse events (SAEs) and serious unexpected serious adverse reactions (SUSARs) to the responsible legal authorities. Annual reports were filed each year, as required. No remarkable safety issues occurred in the RGT.
Accomplishments regarding establishing safety reporting/pharmacovigilance obligations, we have accomplished the following:
• An SAE Form and Medical Monitor second assessment form were designed for use by all trial centers participating in the RGT
• An SAE Form and Medical Monitor second assessment form were designed for use in the Regensburg M reg CTG trial
• VigilanceONE safety database was installed at the University Hospital Regensburg and training conducted by PharmApp. This software is an E2B-compliant safety database that will support data exchange with EudraVigilance (at the EMA) and all national authorities concerned by the planned clinical trials.
• University Hospital Regensburg was registered with EudraVigilance in the training environment (EVTEST)
• Testing of the Gateway connection for the exchange of electronic safety reports with EudraVigilance was successfully completed
• UHREG registered with EudraVigilance in the production environment (EVHUMANable to report trial SUSARs electronically to the EMA
• MedDRA training was undertaken by the Regensburg pharmacovigilance group
• The XEVMPD training course was undertaken at the EMA in London to prepare for entering cell product IMP in the EV Medicinal Product Dictionary
• The Regensburg cell product was entered into the extended EV Medicinal Product Dictionary
• The Regensburg group has exchanged pharmacovigilance experiences with other clinical trial centers responsible for safety reporting
• VigilanceONE has been used to produce line listing summaries to include in annual safety reports (DSURs). For the RGT and the M reg CTG trail, Regensburg has created and submitted a DSUR to the concerned authorities in all participating countries, in line with the applicable regulatory deadlines
A significant achievement has been the development of a robust SAE management system that also incorporates electronic safety reporting functionality. This commitment was supported by investment in the VigilanceONE safety database and in the training of staff members in electronic safety reporting and in eXtended EudraVigilance Medicinal Production Dictionary submissions at the European Medicines Agency (EMA) in London.
Safety - The Cell Therapy Group (CTG) trials
For the CTG trials, a special Safety Advisory Board was established to oversee all of the trials. This board consisted of three members of The ONE Study external advisory board (Sir Peter Morris, Prof. Stuart Knechtle and Prof. Josep Grinyó), who were independent, but yet familiar with the special design and plans of The ONE Study consortium. We decided that it was critical that we have a group of experts to evaluate the safety of all the ongoing CTG trials, since a local data safety monitoring board (DSMB) has been established for each CTG trial; however, as these individual DSMBs do not have a direct communication with each other, only the SAB has an overview of safety in all the combined trials. In case there is a problem with safety in one CTG trials, the SAB would detect this and report their conclusions to the various DSMBs.
Regular meetings of the SAB have taken place approximately every six months during the conduct of the CTG trials. The first meeting of the SAB (via teleconference) was convened in July 2015 and the most recent meeting occurred in June 2017. During these meetings, the SAB evaluated cumulative safety data collected in the CTG database since the 13th May 2014 (date of enrolment of the first CTG patient). In all meetings conducted so far, the SAB has been satisfied with the progress made in the CTG trials and has not identified any major safety concerns. The SAB has been satisfied during the course of The ONE Study for all CTG trials to continue. After each meeting, an official set of meeting minutes were written and approved by the participants, and an official written statement (summarizing the main conclusions) was signed by the three members of the SAB. The central team in Regensburg circulated the official written statement to all CTG groups and asks each trial manager to forward the statement to their local DSMB. In addition to the six-monthly meetings arranged by the central team in Regensburg, the SAB also receives critical safety information (on SAEs and rejection episodes) from the various CTG trials in real-time. Every time an SAE or rejection episode is reported to the central team in Regensburg, the SAB is immediately informed via email and sent an overview table containing a cumulative listing of SAEs and rejection episodes. This allows the SAB to review safety data as it accumulates and detect any significant safety signals as they emerge.
SAB meetings that were held:
- 15th July 2015
- 2nd December 2015
- 21st July 2016
- 30th November 2016
- 27th June 2017
- (planned) 25th January 2018 (next and possibly final meeting-if any trials do add more patients, the meetings will continue)
Summary of safety results for RGT
Through the eCRF systems we have begun to analyze the available safety data from the CTG trials.
The safety data show the normalized frequency of AEs and SAEs in the RGT, compared to the combined events that occurred in the various CTG trials. We obtained a clear result here, where it is shown that the frequency of both AEs and SAEs was essentially identical in the CTG trials, compared to the RGT. This is a very encouraging result, and has provided the basis for new trials that have been initiated by our partners, as described elsewhere in The ONE Study report. Importantly, only 5 SAEs were possibly related to the infusion of the different cell products, and all of these events were easily treated and reversed. Therefore, we conclude that none of the cell therapies has caused a safety concern with regard to AEs or SAEs that were discovered through The ONE Study.
With the general safety of these cell therapies demonstrated by these results, there are two cell therapy products that did raise a cautionary flag. The Ag-specific cell product from UCSF was associated with the occurrence of a dense lymphocytic cell infiltrate shortly after cell infusion. This infiltrate coincided with an increase in creatinine that was successfully treated with standard steroid treatment that is typically used in cases of rejection. It remains unknown whether this infiltrate was due to the infusion of the Ag-specific Tregs. The local DSMB put the trial on hold, but did approve the resumption of the trial once an investigation was completed. The SAB had a similar opinion that the trial could continue as planned. However, it was decided by the local investigators that early application of the Ag-specific Tregs in this protocol might be too risky to continue in this trial; therefore, they decided to halt recruitment of more patients into the trial, and opt for infusion of cells at a later time point in relation to organ transplantation. A second cautionary signal came in the Mreg trial in Regensburg. The second patient treated with Mregs experienced a rejection episode shortly after transplantation and later developed the presence of donor specific antibodies. Since the Mregs are derived from donor cells, there was some concern that the Mregs could have contributed in some way to the donor specific antibody development in this patient. Importantly, no donor specific antibody was detected on the day of transplantation, which suggests that the Mreg cell infusion 1 week prior did not itself induce detectable levels of donor specific antibodies. A complicating issue in this case is that the donor was the husband of the recipient and they have offspring; this raises the possibility that the recipient was presensitized through pregnancy to the donor (husband’s) non-inherited MHC antigens. Therefore, the rejection and development of donor specific antibody could be related to the pregnancy, which is not uncommon. Therefore, the local DSMB reviewed the case and recommended that the study continue, but that this donor recipient combination of husband to wife not be allowed, which could confuse interpretation of future safety data. This recommendation was followed, but the already restrictive exclusion criteria, combined with exclusion of husband to wife donations, significantly narrowed eligible patients available in Regensburg, and has resulted in a long delay since the last patient was treated in the trial. This has prompted consideration as to whether to continue this trial under the existing protocol.
Distribution of SAEs in CTG trials patients vs. RGT patients
An important aspect of cell therapy in general, and therefore of The ONE Study, is that the goal is to reduce the need for pharmacological immunosuppression which leads to side effects that limit outcome success. In The ONE Study, we did slightly lower immunosuppression in the CTG trials by avoiding the use of induction therapy with basiliximab; in addition, there was an option to reduce immunosuppression at 9 months posttransplantation to tacrolimus monotherapy (this was done in about one-third of patients under cell therapy). Therefore, with the lower pharmacological immunosuppression conditions for the recipient, we would hope for a reduction in side effects that are normally associated with general immunosuppression. Indeed, when analyzing the currently available data on CTG trial patients, the frequency of infections dropped dramatically from 42% in the RGT, to 9% in the combined results from the CTG trials. This result provides first evidence of optimism for cell therapy as a means to reduce side effects from general immunosuppression.
The ONE Study has demonstrated that immune cell therapy with the tested T cell and monocyte derived cell products is safe in terms of side effects. Two cell products did raise cautionary flags, but have not been ruled to be unsafe; further testing in patients is needed to reach a final conclusion regarding safety on these products. Moreover, we now have the first evidence that in fact reduction of general immunosuppression via application of cell therapy could be effective in reducing side effects associated with standard pharmacological drugs. These results have given impetus to the further development of these cell therapies for organ transplantation, which is evidenced by the approval of new trials both in Europe and in the USA.
Primary endpoint results
Although the main goal was to determine safety, we also designed the study to get at least an initial set of preliminary data regarding effects of cell therapy on development of transplant rejection and on the immune system. Below the primary endpoint data will be presented, along with some key preliminary findings regarding our immune monitoring program, and on health economics assessments. To perform these analyses data capture systems (eCRF), safety monitoring boards (data safety monitoring board and safety advisory board), and health economics platforms were set up and established for performance before clinical trials were started; therefore, these evaluation aspects were fully in place and on schedule in The ONE Study.
Reference Group Trial Primary Endpoint Results:
As stated above, the RGT trial serves as a control or reference group for the 6 CTG trials. The RGT design was agreed upon by all 8 transplant centers involved in The ONE Study, and received ethical and regulatory approval according to schedule.
The primary endpoint for the RGT was biopsy confirmed acute rejection (BCAR). Eight patients experienced a BCAR, and these results have been confirmed by the central pathologist (Prof. Ian Roberts) in Oxford. There was an uneven distribution of BCAR at the different sites, but this was not entirely unexpected since the development of BCAR is normally only in 1 or 2 patients out of 10. Nonetheless, there were 3 rejection episodes in the 10 patients in Berlin. We are not sure the reason for this high rejection rate, and interestingly there was also a similar rejection rate in the Berlin CTG trial. Indeed, an advantage of The ONE Study design is that we can use both the local data results from the RGT and the results from the whole group to make comparisons; if there are local effects, these can be addressed at least to some degree by looking at both the local and whole data set.
A summary of the primary endpoint (biopsy confirmed acute rejection-BCAR) results are shown below. Note that we have broken the RGT trial down into 2 main groups. The intention to treat (ITT) group included all evaluatable patients, and the per protocol (PP) group included patients that were treated very closely to the planned protocol. The PP population was defined with clear criteria in our trial statistical plan.
RGT results summary:
Based on the literature and our statistical calculations, we expected a rate of BCAR between 2.4-17.6% in the RGT. Therefore, the rejection rate was within our expectations and provides a very useful control for the CTG trials. Regarding the immune monitoring, patient samples in the RGT have been received by the central laboratory at the Charité and there are some key findings to summarize from the RGT: When combining the findings of all RGT patients we did identify a set of biomarkers (e.g. Nav3 gene expression), which allows pre-transplant identification of patients experiencing acute rejection. In addition, we detected alterations in peripheral blood immune cell composition (e.g. % of total B cells, naïve CD4+ T cells) in patients showing a decline in graft function up to 12 months prior to clinical manifestation.
Cell Therapy Group trial Primary Endpoint results
note that the BCAR results for the CTG trials have not all been scored yet by the central pathologist (some only scored by the local pathologist), and therefore must be considered preliminary
• UK trial (Oxford and London): polyclonal Treg-frozen product
All 12 patients were treated with polyclonal Tregs, according to plan, and all patients have completed the 15 month follow-up phase of the trial. In total, 15 patients were recruited into the trial, with 12 actually receiving Treg. None (0%) of the 12 Treg treated patients experienced a BCAR (the BCAR rate in the RGT was 17% in the PP population and 13.6% locally at the two UK centers. The trial has reached completion. Two patients were reduced to Tacrolimus monotherapy after 9 months posttransplantation, and others have been reduced after the 15 month observation period. Note that reduction to Tacrolimus monotherapy was optional within our study protocol.
• Berlin trial: Polyclonal Treg-fresh product
Eleven patients were treated with polyclonal Tregs and all patients have completed the 15 month follow-up phase of the trial. In total, 17 patients were recruited into the trial, with 11 actually receiving Treg. Three (27.3%) of the 11 Treg treated patients experienced a BCAR (the BCAR rate in the RGT was 17% in the PP population and 27.3% locally at the Berlin center. Therefore, although the rate of rejection is relatively high in this CTG trial, it is the same rate as was observed in the Berlin cohort of the RGT. The trial has reached completion. Remarkably, 8 patients were reduced to Tacrolimus monotherapy. It is also notable that of the 11 patients, 2 were from full HLA (6 antigen) matched pairs.
• UCSF (San Francisco) trial: Ag-specific Treg product
Two patients were treated with the UCSF donor-reactive Tregs, and both patients have completed the 15 month follow-up phase of the trial. In total, 5 patients were recruited into the trial, with 2 actually receiving Treg. One of the two patients did show a substantial infiltration with lymphocytes upon kidney biospy one day after Treg infusion, which is consistent with BCAR; however, the central pathologist has not yet confirmed the diagnosis of BCAR. A rise in creatinine also occurred in parallel with the cell infiltration and the patient did respond to steroid treatment of the condition. Nonetheless, there was enough concern by the local investigators that infusion of donor-specific Treg under this protocol is not being continued, even though the data safety monitoring board did not suggest to discontinue the study. Therefore, the decision to stop recruitment of more patients was made by the local investigators. No patients were reduced to Tacrolimus monotherapy.
• MGH (Boston) trial: Ag-specific Treg product
Three patients were treated with Ag-reactive Tregs, and all 3 patients have completed the 15 month follow-up phase of the trial. In total, 5 patients were recruited into the trial, with 3 actually receiving Treg. None of the 3 Treg treated patients experienced a BCAR. Two of the patients were reduced to Tacrolimus monotherapy at 9 months according to the protocol option, and the third patient was also reduced to monotherapy, but only after the 15 month follow-up period. The investigators elected to not recruit more patients into the trial because the yield of Ag-reactive Tregs with their manufacturing protocol was acceptable for infusion, but low. They decided to establish another manufacturing procedure that will bring a greater yield of Ag-reactive Tregs and start a new trial. Indeed, the investigators now have funding to conduct a new trial using the “improved” cell manufacturing procedures.
• Nantes trial: tolDC product
Eight patients were treated with tolDC, and 5 of the 8 patients have completed the 15 month follow-up phase of the trial; 3 remain in follow-up that will extend past the end of The ONE Study. In total, 11 patients were recruited into the trial, with 8 actually receiving tolDC. Two (25%) of the tolDC patients have experienced a BCAR, thus far; for reference, the PP RGT patients had a 17.0% BCAR rate, and none of the Nantes RGT patient cohort experienced a rejection episode. None of the patients have been reduced to Tacrolimus monotherapy. The investigators currently have their trial on hold, pending additional funding for treating additional patients.
• Regensburg trial: Mreg product
Two patients were treated with Mreg, and both have completed the 15 month follow-up phase of the trial. In total, 8 patients were recruited into the trial, with 2 actually receiving Mreg. Notably, one of the two patients experienced a BCAR. There were two problems with this trial that limited the number of patients able to be treated. First, there were problems with the cell manufacturing procedures that resulted in several of the recruited patients not being treated. These manufacturing problems were solved over the course of approximately 1 year, but then there was an issue at the site regarding patient eligibility; while over 20 patients were screened for inclusion into the study in 2017, none of them fulfilled the inclusion criteria. Neither of the two patients have been reduced to Tacrolimus monotherapy. The study has effectively been halted at this point, due to funding issues, since The ONE Study funding period has ended. Issues regarding safety of using a “donor-derived” cell product were addressed in one treatment case with BCAR and donor-specific antibody, as described in the deliverable, D4.1. Although the safety monitoring board has given the trial a positive evaluation to go forward, lack of additional funding limits the possibily to treat more patients.
Summary of CTG trial results as a whole:
• 60 living-donor kidney transplant recipients enrolled
• 45 underwent transplantation
• 38 received cell therapy
• 22 withdrawn prior to receiving cell therapy
• Primary endpoint (BCAR) based on local biopsy pathology
• Rate of BCAR (local pathology) in all CTG cell-treated patients = 15.8 %
Below is a summary of the inclusion and primary endpoint results within the individual CTG trials. Notably, the central pathologist has not yet confirmed all cases of BCAR, but will complete this work by January 2018. It is worth pointing out that the Oxford/London UK trial is one clinical trial; therefore, the primary endpoint results are combined. All of these trials are either completed with recruitment, or have been stopped due to the reasons described elsewhere in the report. The most promising trial results came from the UK polyclonal Treg trial, where they treated 12 patients, none of which experienced a BCAR. In response to these positive results, the UK group has just recently obtained funding from Medical Research Council of the UK to perform a follow-up trial in kidney transplant recipients; this trial is called “The TWO Study”, and will treat around 40 patients. The Boston MGH group has also just received funding from the NIH to perform a new trial in transplant recipients based on an improved yield manufacturing protocol for Ag-specific Tregs. The UCSF group has also received additional funding to perform new trials using their cell product in renal transplant recipients (cells given after approximately 6 months after transplant) and in liver transplant recipients after more than one year after transplantation.
CTG results summary:
Notably, 12 patients were reduced to receiving only Tacrolimus monotherapy starting approximately 9 months after transplantation. This is an overall rate of conversion of 32%, among cell therapy recipients. There is an uneven distribution of these patients among the CTG trials, due to the fact that monotherapy conversion was optional at 9 months in the protocol. Nonetheless, approximately one-third of the combined cell therapy patients are on a single immunosuppressant. This proportion increases when including patients that were weaned to monotherapy after the 15 month observation period. Importantly, none of the patients on monotherapy have experienced a BCAR to date.
The ONE Study serves the important purpose to show that monotherapy converstion is a next stepping stone for future cell therapy protocols. Until these results became apparent, reduction down to monotherapy may have been considered not advisable by some nephrologists. We believe these results set an important precident for future clinical trials aiming for monotherapy using a cell therapy approach.
Immune monitoring (IM) program
To reveal differences between the patients enrolled into the RGT and into the CTG trials, a specific IM program was developed by which the safety (e.g. EBV viral load), pharmacodynamics (DNA microarray) and efficacy (e.g. flow cytometry of leucocyte subsets) parameters are recorded. The central IM group in Berlin has prepared and distributed the SOPs for blood, serum collection, blood sample preparation and assay performance. Furthermore, antibody panels for flow cytometry were designed, tested for their performance in the central immune monitoring lab and also at all involved clinical trial sites with four interlab comparisons. Beckman-Coulter Life Sciences has provided eight 10-color Navios cytometers (one per site). Training for running the Beckman Coulter Navios flow cytometer and for procedures regarding our custom antibody panels was done personally at each site by Matthias Streitz (Charité).
Beckman-Coulter Life Sciences collaborated with the Charité on the design and development of six immune monitoring panels for the detection of T, B, NK, monocytes, dendritic cells and associated subsets by flow cytometry. Validation of the panels lead by the Charité with the support from Beckman-Coulter Life Sciences has demonstrated the sensitivity and robustness of the panels and the very effective standardization of the flow cytometry workflow (Streitz et al.; Transplantion Research 2:17, 2013). Throughout the study Beckman manufactured and supplied a total of 15000 tests of six 7- to 9-multi-color panels. To date more than 12000 tests were already shipped to The ONE Study consortium. Since October 2014, the T, B, NK and the B cells panels are available as commercial products by Beckman-Coulter. Two additional panels, i.e. T cell subsets and DC panels, strongly inspired by The ONE Study panels were launched in December 2014; with two more panels being developed for launch. For all these panels a new strategy to set-up the flow cytometer and to establish compensation has been implemented to facilitate a large adoption by the scientific community. Additionally, Beckman-Coulter has been supporting other studies such as the Canadian National Transplantation Research Program to adopt The ONE Study panels. All The ONE Study panels can + cv be ordered commercially through the Beckman Coulter Custom Design Service in dry and liquid format. Our consortium and the Beckman Coulter group have shown that the panel is marketable and considerable efforts have been made to bring this forward to the research community. Some of the developed flow cytometry panels are now commercially available as “off-the-shelf” products (by Beckman Coulter Custom Design Service) for the whole research community in dry and liquid format.
Immune monitoring data for the CTG trials
Analyses of the CTG trials are still ongoing, but preliminary results indicate that Treg transfer can ameliorate alterations in immune cell composition in a dose-dependent manner. However, the final immune monitoring results for the CTG trials will not be possible until all of the data from the 6 trials is monitored completely, which we expect to be done in January 2018. Then, the analyses can be performed for each of the trials to look for effects of the various cell products on the immune system in kidney transplant recipients. It is critical that all the analyses and bioinformatics be performed on a complete set of monitored data (January 2018), since this is the only way to assure the fair comparison of results from the trials. Therefore, we will report on the immune monitoring data within the publications that will result from The ONE Study in 2018 (see deliverable D8.8 report).
Conclusions regarding primary endpoint outcomes in the RGT and CTG trials
The results from The ONE Study RGT have provided an excellent comparator group for our CTG trials. With the numbers of patients in these phase I trials, we can only make substantial statements about safety, which has been addressed and summarized in deliverable D3.2.
Regarding signals of efficacy, we cannot make any strong statements because of the small numbers. However, we can say that overall, cell therapy has not generally resulted in higher numbers of BCARs, with between 12.1-17.0% BCARs in the RGT trial (depending on whether using data from the IIT population or PP population), and 15.8% among all of the CTG trials. Individually, some trials experienced higher or lower rejection rates, but there is little that can be concluded with such small numbers of patients in these safety trials. Nonetheless, the most promising trial signal came from the UK polyclonal Treg trial, where they treated 12 patients, none of which experienced a BCAR episode; this is an impressive result, since the UK sites had a BCAR rate of 13.6% in the RGT and the overall BCAR rate was approximately the same in the RGT. In response to these positive results, the UK group has just recently obtained funding from Medical Research Council of the UK to perform a follow-up trial in kidney transplant recipients (The TWO Trial). The Boston MGH group has also just received funding from the NIH to perform a new trial in transplant recipients based on an improved yield manufacturing protocol for Ag-specific Tregs. The UCSF group has also received additional funding to perform new trials using their cell product in renal transplant recipients (cells given after approximately 6 months after transplant) and in liver transplant recipients after more than one year after transplantation. Therefore, the safety profile of cell therapy has proven to be acceptable thus far, and the evidence does not suggest a higher risk for rejection, even if immunosuppression is lowered with cell therapy to tacrolimus monotherapy. Some caution will need to be taken when moving forward the use of the cell products from UCSF and Regensburg, as indicated in this report. Notably, a new clinical trial that has been approved at the University of Pittsburgh essentially uses the same clinical trial protocol as that used in The ONE Study and the Pittsburgh group was fully informed by The ONE Study consortium. In fact, the Pittsburgh group is using a donor-derived dendritic cell product that is of the same categorical type as the Mregs used in Regensburg (using also similar doses and timing of cell infusion); the Regensburg group has closely advised the Pittsburgh group about lessons learned from their trial regarding the use of donor cells. Indeed, the Regensburg group shared protocols with the Pittsburgh group. Therefore, The ONE Study has formed the main information base for which immunoregulatory cell therapies are now being taken forward into new trials in organ transplantation. The experience has been invaluable to the international cell therapy and transplantation reseach communities.
Health economics report:
The aim of the health economics analysis is to use data from the reference and cell therapy group trials in The ONE study to evaluate the potential cost-effectiveness of cell therapy as a new therapeutic technology in renal transplantation
The primary perspective of the economic evaluation was that of the healthcare system, with primary, outpatient, emergency and inpatient care costs included. As a result, costs associated with premature death or time off work (i.e. loss of productivity) and informal care (i.e. the time devoted by friends and relatives providing unpaid care to patients) were not included.
Quality of life
Generic health-related quality of life (HRQoL) was measured using the Euroqol-5 Dimensions-5 Levels (EQ-5D-5L) questionnaire. Patients were asked to think about their health in the day they are completing the questionnaire and report on any problems (none, slight, moderate, severe and unable/extreme) on 5 attributes (mobility, self-care, usual activities, pain/discomfort and anxiety/depression). Patients are then required to rate their health using a 100-point visual analogue scale (VAS; 0=worst health you can imagine to 100=best health you can imagine). The EQ-5D is a standardised measure of health providing a simple generic measure of health for clinical and economic appraisal. The EQ-5D was administered at visits 1, 4, 7, 9 and 10 for all the trials.
In addition, generic HRQoL was measured using the 36-Item Short Form Survey (SF-36). The SF-36 consists of 36 questions, measuring eight health aspects: vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning and mental health. The SF-36 was administered at visits 1, 7 and 10 in the various trials.
As part of the healthcare system perspective adopted we included the following healthcare resource categories over the 60-week follow-up:
• Community care costs, including visits to a: general practitioner (at surgery, home or through the telephone); nurse (at surgery or at home); physiotherapist; occupational therapist; psychologist; and counsellor; and
• Hospital care costs, including: visits to accident & emergency (A&E); ambulance use; and length of stay in hospital including stays in an intensive treatment unit (ITU). We also included the costs associated with the kidney transplant, and any subsequent procedures.
Information on the resources used as part of the initial hospitalisation for renal transplantation was obtained from a review of medical/administrative patient records by trial staff at each of the centres in The ONE Study, with a study form developed to help site staff collect this information. Resource use once discharged from hospital was collected using patient questionnaires administered at visits 4, 7, 9 and 10. As an aide memoire, patients were also provided with a resource use log designed for them to fill in every time they had a contact with the healthcare system.
In order to value resource use information collected from different study sites and countries, the same unit costs will be used to value all resource use. For this study, unit costs from the United Kingdom will be applied.
Unit costs for consultations with GPs and nurses will be obtained from the Personal Social Services Research Unit’s Unit Costs of Health and Social Care publications for 2015 and 2016.
For all other NHS contacts, unit costs will be derived from the NHS National Schedule of Reference Costs 2015 to 2016. For outpatient visits, we will use the weighted average of all consultant led non-admitted face to face attendances, either first or follow-up . For physiotherapists, occupational therapists and psychologists we too will use the weighted average of all consultant led non-admitted face to face attendances for each of these three therapists. In the absence of specific unit costs for counsellors, we will assume that these will be the same as for a psychologist. For visits to accident and emergency (A&E), we will use the weighted average of all emergency medicine contacts, excluding dental care and patient dead on arrival. For ambulance transport to A&E, the unit cost of a call to the emergency services will be included as well as that for ambulance transport.
For the initial hospitalization, where the patient underwent kidney transplantation, the unit costs per day case/day in hospital is derived from the schedule of NHS reference costs.(6) Using the reasons for subsequent hospitalisations reported by patients in the resource use questionnaires, we will obtain diagnosis and procedure codes. These will then be translated into a Health Resource Group (HRG) using the HRG4+ 2015/16 Costs Grouper (NHS Digital). Each HRG will then be linked to a series of elective, non-elective and day case costs obtained from the 2015/16 schedule of reference costs.
Details of the medications patients received after their kidney transplantation will be obtained from trial records. The cost of these medications will be obtained from the British National Formulary (BNF), using generic drug prices if a particular medication has gone off-patent.
Quality of life
For each follow-up, we evaluate the scores for each of the eight health attributes in the SF-36. For each attribute the scores are the weighted sums of the questions in each section. We then evaluate the summary score of physical QoL (Physical Component Summary; PCS) and emotional QoL (Mental Component summary; MCS) by estimating the mean average of all of the physically relevant questions and of all of the emotionally relevant items.
At each follow-up, responses to each of the 5 questions in the EQ-5D-5L are presented. As recommended, EQ-5D-5L responses are converted into utilities using the validated mapping function to derive utility values for the EQ-5D-5L from the existing EQ-5D-3L.
SF-36 scores, utility values and EQ-VAS scores at each follow-up will be presented as means alongside standard deviations (S.D.). Differences in responses to each of the 5 questions in the EQ-5D-5L between the two treatment groups will be assessed using chi-square tests. Mean differences across the two treatment groups will be presented alongside 95% confidence intervals, with statistical significance assessed using two-sided t-tests. Differences between 60-week and baseline scores and utilities will be assessed within- and across-patient groups and will also be compared using two-sided t-tests. For utilities and EQ-VAS scores the results will be presented for complete cases only (i.e. patients who had complete EQ-5D-5L and EQ-VAS information).
Survival will be estimated using the Kaplan-Meier survival function 60 weeks post-enrollment in the study. A quality-adjusted survival curve will be generated by plotting, against time, the product of the mean utility of patients living at time t and the probability of surviving to time t, with t being set to represent the time at each visit, up to 60 weeks follow-up. The area under this quality-adjusted survival curve will then give the mean quality-adjusted survival in each treatment group. Utility will be assumed to change linearly between each follow-up, rather than changing at the midpoint between follow-ups or being maintained from one follow-up to another.
For each treatment group, results will be reported as QALYs with 95% CIs calculated non-parametrically from 1000 bootstrap differences. Mean QALY differences between the two patient groups will also be presented with 95% CIs estimated using the 1000 bootstrap differences. Results will be presented for the whole patient sample (i.e. were patients who withdraw from the analysis will be treated as censored and missing utility estimates were assumed to be the same as the mean for that treatment group) and complete cases (only patients with complete EQ-5D responses or those who died will be included).
Resource use and costs
All contacts with the healthcare service will be reported as means (S.D.). Except for days in hospital, statistical differences in resource use between the two patient groups will be evaluated assuming a poisson distribution. Days in hospital between both groups will be assessed using a Student’s two sided t-test. Results are presented for available cases, in which resource use at each follow-up will be presented, and for complete cases, in which resource use for patients with complete resource use information over the 60 weeks will be reported.
All costs will be presented as means (S.D.) and reported in Euros (€) using the rate of purchasing power parties. Differences between the two patient groups will be assessed using a Student’s two-sided t-test. Costs will first be presented for available cases, in which costs at each follow-up will be reported, and then for complete cases, in which costs for patients with complete cost data over the 60 weeks will be presented.
In order to estimate the potential cost-effectiveness of cell therapy as compared to current practice, we will undertake a complete case analysis including only patients who had complete cost and EQ-5D utility information. We will carry out an incremental analysis, with the mean cost difference between the cell therapy and current practice divided by the mean QALY difference to give the incremental cost-effectiveness ratio (ICER). Of course, this estimate will not include the costs of cell therapy, which at this very preliminary stage of production are still difficult to determine. As a result, based on the results of the ICER, we will perform threshold analyses to assess at which cost would cell therapy look economically advantageous (i.e. at which cost would cell therapy yield an ICER of less than £20,000 to £30,000, €17,500 to €26,000, representing the current cost effectiveness threshold in the UK).
To calculate the confidence interval around the ICER we will use the non-parametric percentile method, using 1,000 bootstrap estimates of the mean cost and QALY differences. We will then use a modified version of the cost-effectiveness acceptability curve to show the probability that cell therapy is cost-effective at the €17,500-€26,000 cost-effectiveness threshold at different costs for the cell therapy.
Reference case - after multiple imputation of missing data
Multiple imputation will be used to impute missing cost and utility values. As recommended best practice, imputation was implemented separately by randomised treatment allocation. Costs will be imputed at the most disaggregated level at which the model would converge.
Preliminary results from the Reference Group Trial
For patients in the Reference Group Trial (RGT), mean utility was comparable at baseline and four weeks (0.81 vs. 0.80). However, by 24 weeks, utility had increased to 0.88 by 24 weeks. By 60 weeks, utility was 0.86. The graph obtained shows how, after imputing for missing cases, utility estimates were comparable between complete cases and after multiple imputation. Overall, over the 60-week follow-up (i.e. 1.15 years) the QALYs gained were 1.05.
Figure 1. Mean utility over 60-week follow-up
The mean days in hospital for the hospitalization in which the patient had the kidney transplant was 11.4 days (S.D. 7.6). However, this mean length of stay varied widely across sites and countries (mean days in hospital varying from 5 to 23).
However, the most critical analysis will compare the RGT data with the cell therapy group trials as a whole and individually. Once the cell therapy group trials are completely monitored (January 2018), and analysis as described herein will be initiated.
WP5: Comparative characterization of cell products
An objective of The ONE Study consortium was to gain a better understanding of the similarities and differences between different cell therapy products. To define the differences more precisely, with the final aim to be able to design new cell products with added potential, we set up a workshop in Regensburg in 2012 to produce antigen presenting cells of interest and analyse differential phenotypic and functional characteristics. The workshop in Regensburg was a complete success, and the data are still being analysed with the aim of publishing the results in 2018. Because of the substantial investment in time to do this comparative analysis in a workshop form, and limited remaining available funds for this purpose due to the high costs of cell therapy manufacturing being experienced by all the sites, we decided to not go ahead with a formal workshop on Tregs. Nonetheless, partners within our group are also part of an EU funded COST initiative (BM 1305) and are working together to investigate, discuss and compare immunosuppressive cell products. Together with BM 1305 we recently published an article in Science Translational Medicine (7:304ps18, 2015) on Treg types and therapy. It is also important to mention that our various partners have published several manuscripts that describe both T cell and monocyte-derived cell products, GMP production and details of immune regulatory cell properties (Blood Adv 1:947, 2017; Transplantation 101:2731, 2017; Transplant Int 30:765, 2017; Peer J 4:e2300, 2016; Frontiers Immunol, in press; Am J Transplant 16:2187, 2016; Am J Transplant 13:3010, 2013). This is a major accomplishment of The ONE Study and we expect this information will be extremely helpful to our groups and other investigators interested in using cell therapy in the future.
Another objective of WP5 was to standardize assays for release-criteria of cell products across all the partner sites. However, since the cell products were manufactured in different countries that have their own regulatory bodies, the testing requirements are slightly different; centralisation of testing for cell manufacturing was not practical to implement. Therefore, we concluded, after receiving feedback on all of our cell products from our respective competent authorities, that we did not feel it practical to put effort into harmonising the manufacturing requirements in this ground-breaking cell therapy study.
After The ONE Study subgroup meetings, and the meeting with the EMA, our group has a better understanding of what can be accomplished in this respect. Openly sharing testing methods and strategies can be mutually helpful within our group of investigators, and we strive to maintain effective communication lines and exchange data for this purpose. Above, we have listed a number of recent publications where we describe in detail our experience and give insight into cell types and manufacturing methods. As a group, we now realise that since the regulatory specifications are continuously being modified to optimise the cell products, it is very difficult to pin-point which quality control methods are best to standardise.
WP6: New cell populations with tolerogenic potential
Several groups have made steady scientific progress on cell product development regarding Treg and tolDCs, as well as NK cells.
At the KCL site, substantial experimental work has been advanced:
(1) Compared drugs such as rapamycin and all-trans-Retinoic acid (ATRA) to obtain the most suppressive and stable (lack of conversion to IL-17 producing cells) Treg lines using research grade reagents. They have established that although ATRA treatment of Tregs increases their function, their stability is altered. ATRA-Tregs produce spontaneously high levels of IL-17 and IFN-γthat can be overcome by adding rapamycin to cultures. Moving to a humanised mouse XenoGvHD model, they have confirmed in vivo that ATRA-treated Tregs produced IL-17 while rapamcyin- and ATRA+rapamycin-treated Tregs were not affected by the inflammatory milieu and showed a stable phenotype. One of the mechanisms for the inhibitory function of rapamycin towards Treg conversion has been established. Further analysis of the effect of drug treatments on Treg subsets showed that rapamycin expands preferentially the most “naïve” and stable Treg subpopulation. The manuscript summarising these results was published in Haematologica (98:1291, 2013). They have also studied the gene transcription profile of rapamycin-, ATRA- and ATRA+rapamycin-treated Tregs and compared their fingerprint with untreated Tregs. They have identified a list of differentially expressed genes that describe the gene signature of treated Tregs. These fingerprints include only a few genes previously described as substantially regulated by either rapamycin or ATRA (e.g. mTOR activity, retinoic acid), so they represent unique transcriptional changes in Tregs that characterise functional and stable Treg phenotypes. Of particular interest, several of these genes can be targeted with available drugs, and will be investigated in their own right for a role in Treg stability/function as an alternative to conventional treatments. More recently the genetic profile of Tregs from patients with systemic lupus erythematosus has been investigated. The hypothesis is that the decreased function of Tregs in these patients can be explained with an altered genetic profile that can be ‘corrected’ by culturing Tregs in the presence with rapamycin. They are now waiting to complete the genetic profile of Tregs from systemic lupus erythematosus freshly isolated and after expansion in vitro with rapamycin.
(2) The group has characterised Treg subtypes in dialysis patients and has established the optimal culture conditions (media, IL-2 dose, rapamycin dose, expansion beads, plastic) for the production of large numbers of functional and stable clinical grade Tregs using blood from healthy controls and dialysis patients. More recently they have studied the stability of the Treg lines obtained from hemodialysis patients and although they converted to IL-17-producing cells in the absence of rapamycin, expansion in the presence of rapamycin inhibited their conversion. A related manuscript was published in 2013 in Clin J Am Soc Nephrol (8:1396-1405). The work on the stability of kidney transplant patients has been extended to patients with liver diseases with similar results. The first manuscript (just resubmitted after revision) describes the clinical protocol for the expansion of Tregs from patients with liver cirrhosis and in the second manuscript they have presented the investigation of the mechanisms behind the defect in Treg function that they have identified in patients with liver cirrhosis (manuscript is in preparation).
(3) The group has identified the population of Tregs that is responsible for the conversion of Tregs to IL-17 cells. They have also established that Treg conversion correlated with the presence of specific pro-inflammatory cytokines such as IL-1 beta and have shown that it is STAT-3-dependent. In the same study it has been established that the expression of CD161 identify the plastic Treg population. A manuscript was published in 2013 in the European J Immunology 170:300. The analysis of the function of CD161+ Tregs has been the main project of a PhD student. He has investigated in vitro and in vivo the function of CD161+ Tregs compared with CD161- Tregs (a manuscript is in preparation).
(4) The group has investigated which Treg subpopulations from patients should be enriched, if any, to further improve the function and stability of Tregs. The group has recently investigated the difference in the suppressive ability of Tregs (CD45RA+ and CD45RA-) in patients with Crohn’s disease. The group have established that freshly isolated Tregs CD45RA+ were more suppressive and more stable compared to CD45RA-. This work has been recently published in Gut (65:584, 2015). However the heterogeneity of Tregs in the peripheral blood is even more complex and the KCL group has recently published a manuscript in which the Treg phenotypic complexity has been evaluated using the CYTOF (Mason et al. J. Immunol.).
(5) KCL has characterised a quick assay to measure suppressive activity of Treg within six hours by using the up-regulation of molecules like CD40L on responder T cells. This method was published in 2013 in Blood (119:e57-66).
(6) The KCL group has focused also on the effect of the three immunosuppressive drugs used in the clinical protocol on the function of Treg lines in vitro. The study has investigated the effect of immunosuppressive drugs tacrolimus, mycophenolate and prednisolone on rapamycin-treated Treg preparations (the same Treg preparation that are injected into patients). Results showed that the immune-suppressants do not affect Treg functions, phenotype and the expression of molecules essential to traffic into inflammatory sites. As expected, proliferation and viability of Tregs was considerably affected by high doses of drugs. However, after drug withdrawal, Treg could promptly recover their proliferative capacity. This work has been moved to a humanised mouse model. Viability, stability and function of rapamycin-treated Tregs have been tested in the presence of the same immunosuppressive regimen used in the clinical trial to treat renal transplant patients. The results in vivo confirmed the findings in vitro showing that concurrent immunosuppressive therapy during Treg administration does not affect the phenotype, stability or function of Tregs. A manuscript has been published recently in Heamatologica (Jan 2016).
(7) The KCL group is also testing the use of chimeric antigen receptor donor specific Tregs, and their potential for inhibiting transplant rejection in a humanized mouse model.
Development work in Berlin:
The group from Charité has also been successful in establishing the direct isolation of human alloantigen-specific natural Treg from cell mixtures after stimulation with appropriate target cells. Further studies have been initiated in order to adapt the system to be used in clinical settings.
Development work at the Nantes site:
Studies have been performed to characterise and optimize human TolDCs. Results obtained from the microarray assay realised in the framework of the APC workshop (Regensburg, 2012) have been analyzed. The transcriptome of our TolDCs was compared with the transcriptome of classical DCs (immature) and macrophages to determine if TolDCs might be considered as DCs or macrophages. Different markers of four major properties of APCs were analysed: the antigen processing/presentation pathway, chemotaxis, endocytosis and classical markers of DCs and macrophages. Interestingly, the transcriptome of TolDC is nearer to classical DCs than macrophages for markers of the processing/presentation pathway. To confirm these results, the group continues to work on an in vitro cross-presentation assay with another INSERM unit in Nantes specialized in the study of immunogenic DCs in cancer. Preliminary results indicate that TolDC are able to successfully cross-present antigens to T cells.
Like in animals, human TolDCs are resistant to maturation in vitro upon LPS±IFN-gamma stimulus. This ability of TolDC to resist stimulation by other TLR ligands was tested. These experiments showed that TolDC do not over-express maturation markers after TLR ligand stimulation. Furthermore, no secretion of IL-12 was detected upon stimulation, whereas IL-10 production was detected with some TLR ligands. Moreover, they are investigating mechanisms by which human TolDC suppress T lymphocyte proliferation. So far, the use of inhibitors for the immuno-modulatory molecules HO-1, IDO and iNOS and the study of apoptosis do not provide evidence of a predominant mechanism. In parallel to these in vitro experiments, in vivo experiments using humanised mice (NOD.SCID.gammaRc-/-) have been performed. Results show that TolDC delay the development of GVHD in humanised mice. Furthermore, some experiments are ongoing to determine the survival and homing of TolDC in humanized mice in collaboration with the LU team (WP2). So far, TolDC are detected until day 14 post-injection (no further time points were studied) mainly in liver, lung and spleen.
The Nantes group has investigated more precisely the in vivo behaviour of ATDCs following their injection in humanized mice. They showed that these cells are able to survive up to 14 days following injection with a preferential migration to the spleen. Furthermore, they observed that injected ATDCs are highly viable after injection and preserve their immature phenotype. They confirmed that ATDCs are able to significantly delay GVHD development and are currently studying the in vivo mechanisms of ATDCs in this model. They have confirmed that ATDCs are able to cross-present antigens in vitro efficiently and are working on the analysis of ATDCs effects on T cells. The goal is to develop a more effective and specific ATDC for therapeutic use.
Suppressive NK cell development work in Regensburg:
This group has previously published that NK cells can both damage allografts under inflammatory conditions and, contrarily, promote transplant tolerance by eliminating graft-derived donor DCs. Moreover, they have discovered that NK cells are functionally heterogeneous and that CD27high NK cells preferentially respond to the cytokine IL-15 and that they develop a mature CD27low phenotype upon adoptive transfer. CD27low NK cells, on the other hand, maintained a stable phenotype under cytokine stimulatory conditions. Importantly, studies in humanized mice that they have published reveal that the pool of IL-15-expanded NK cells in those mice consists mainly of functional CD56+CD27low NK cells, indicating that mature human CD27low NK cells properly differentiate and expand upon stimulation with IL-15. The functions of those NK cell subsets are regulated by hallmark T-box transcription factors. CD27high NK cells are contingent on Eomesodermin (Eomes,), whereas CD27low NK cells are critically dependent on T-bet. Importantly, CD27low NK cells are completely absent in T-bet-deficient mice. Thus, they created T-bet deficient mice on a Rag knockout background that lack differentiated CD27low NK cells, as well as T cells. Those Rag-T-bet double-deficient mice represent an excellent preclinical model to study the role of CD27low versus CD27high NK cells in models of allograft rejection and tolerance induction. Using an adoptive transfer model, they have found that Rag-T-bet double-deficient mice fail to clear MHC-mismatched target cells, indicating that CD27low NK cells represent a potentially pro-tolerant subset of NK cells as their clearance of donor-derived MHC-mismatched antigen presenting cells may prevent priming of alloreactive recipient T cells. This study was published (J Immunol192:1954, 2014). Based on these findings, the group has applied this concept to transplant models, and has found that CD27low NK cells prolong allograft survival by controlling alloreactive CD8+ T cells (Transplantation 99:391, 2015). Through this experimentation, the group is considering the potential for manufacturing for CD27low NK cells for therapeutic use.
Most recently, this group has been studying NKp46+ cells as major effector cells in the pathogenesis of hepatic ischemia reperfusion injury (IRI). In these studies they examined the role of NK22 cells in IRI in transplantation, particularly with respect to regulation by the transcription factor ROR-gamma-t (RORγt). Rag KO mice were used for these experiments because they possess fully functional NKp46+ cells, and Rag-common-γ-chain-double-KO (Rag-γc-DKO) control mice were also used because they lack T, B and NKp46+ cells. They found that Rag-γc-DKO mice lacking NK22 cells show more severe levels of hepatocellular damage when compared to both Rag-RORγt-reporter and Rag KO mice that possess NK22 cells. They found that Rag-RORγt-reporter and Rag KO mice undergoing IRI expressed high protein levels of both IL-22 and GFP (RORγt), suggesting a protective role for RORγt+ NK22 cells in IRI. Therefore, they hypothesize that RORγt critically protects from IRI through the induction of hepatic NK22 cells and have published the results supporting this theory (J Hepatol 64:128, 20165). In conclusion, the investigators suggest that RORγt+ NK22 cells play an important protective role in IRI in mice, and could potentially be targeted for therapeutic use. These continuing studies again support the idea of eventually developing an NK cell product with immunosuppressive therapeutic potential.
The expected outcome of this comprehensive program was that the most promising haematopoietic cell therapy products were tested in clinical trials on humans, with the expectation that cell therapy can ultimately reduce the need for immunosuppressive drugs in organ transplant recipients. Importantly, results of this study already show that this goal has been achieved. First, testing of 6 different cell products now provide a safety database for future clinical trials using immunosuppressive cell products in kidney transplant recipients. The data show that cell therapy in general is safe, which is a major conclusion of The ONE Study. Second, regarding cell products with promise, the UK trial using polyclonal Tregs achieved the most impressive results. The 12 treated patients in the UK trial did not experience any transplant rejection episodes, even with a reduction in immunosuppression. Indeed, immunosuppression was reduced substantially in The ONE Study cell therapy group trials in one-third of all patients, and all others had a reduction by omitting induction therapy with basiliximab. Therefore, immunosuppression could be reduced to at least some degree in all the cell therapy treated patients. Third, through the reference group trial and cell therapy group trials, we are able to provide the first evidence that reducing immunosuppression in kidney transplant recipients has beneficial effects. More specifically, as a whole, those patients treated with cell therapy showed only about a 10% chance for developing a serious posttransplant infection, compared to nearly 50% in the reference group trial. Fourth, the strategy to reduce immunosuppression safely to monotherapy was successful and is serving as a model for future cell therapy trials worldwide. In summary, The ONE Study trials show that cell therapy has promise to reduce the need for general immunosuppressive drugs in kidney transplant recipients, and that this reduction could indeed reduce side effects associated with organ transplantation medicine. If this result is confirmed in future trials, patients will not only benefit from having fewer complications posttransplantation, the societal costs for organ transplantation could be reduced. Furthermore, the data obtained in The ONE Study has already been used to support the approval of several new trials, including among others - The TWO Study (in the UK using polyclonal Tregs), thus accomplishing the goal of identifying the most promising new cell therapies that should be tested further in transplant recipients. For these reasons, The ONE Study has proven to be a pioneering effort impacting the course, and future, of cell therapy in organ transplantation.
The ONE Study consortium has put much effort into communicating our study to the medical community, and otherwise, publically. Indeed, The ONE Study has received much attention abroad, and is considered one of the most highly anticipated and regarded set of coordinated trials in the field of organ transplantation. The coordinator has been invited to present The ONE Study concept to many medical societies and organizations internationally. To show the magnitude of ONE Study presentation coverage, the coordinator has been invited to at least 14 events over just the past year to give updates. Among the variety of organizations are the Non-human Primate Transplantation Tolerance Workshop group (NIH sponsored), the American Society of Nephrology annual meeting, Club de la Transplantation (Lille, France), Type 1 Diabetes TrialNet meeting, Advancing Transplantation meeting (Stockholm), meeting of the competent authorities on organ transplantation (European Commission, Brussels), the International Pediatric Transplant Association, the International Pancreas and Islet Transplant Association, Third International Workshop on Clinical Tolerance (held in Stanford, CA), Cellular and Molecular Mechanisms and New Therapeutic Concepts in Transplantation (held in Hannover, Germany – The ONE Study was a co-sponsor), and the American Society of Transplantation and European Society of Organ Transplantation meetings. These are just to name some of the presentations, not to mention the presentations given by partners other than the coordinator. There has been an extraordinarily strong effort to publicize The ONE Study effort and to learn from this experience, with the name of The ONE Study becoming a “brand” known throughout organ transplantation. A ONE Study PowerPoint presentation has been made available to all members of the consortium for their use at meetings where they are speaking on the topic.
With completion of the Reference Group Trial (RGT) and the completion of trials within the Cell Therapy Group (CTG) trials in January, 2018, we are now planning to publish our main results. A plan for these publications has been formulated. Two levels of publications will be sought. One main publication is expected to describe The ONE Study concept and present the first results of RGT and some combined results of the CTG trial data. This manuscript will not contain data of the individual CTG trials, but will present these results only as a whole. We have contacted the editor-in-chief of the Lancet (Richard Horton), and he is interested in such a manuscript and has agreed to handle this manuscript personally. Our current schedule is aimed towards submitting this manuscript by March 2018. The second level of publication is the submission of parallel manuscripts on each of the CTG trials. Our aim is to standardize these manuscripts and submit the 6 articles to a Transplantation specialty journal. These manuscripts will be prepared also at the beginning of 2018 and will be submitted together after the main manuscript to the Lancet is finished.
Cooperations with other projects/programs
The ONE Study group has been working in close connection to the European Union COST action (BM 1305) entitled: Action to Focus and Accelerate Cell-based Tolerance-inducing Therapies. Several members of The ONE Study consortium participate in the BM 1305 COST action. This cooperation has now led to the plan by Dr. James Hutchinson (member of both The ONE Study and COST action) to submit an application for a new Innovative Training Network (call: H2020-MSCA-ITN-2018) with the aim of establishing a training school for regulatory cell therapy in Europe. This new proposal involves members of both groups and incorporates new industrial and academic partners to train a new generation of researchers from bench to bedside on regulatory cell therapy applications. We feel this is an excellent opportunity to make use of our existing leadership on this topic, which has developed from years of funding through cooperations within the European Union funding programs.
As already stated, other funding organizations have been interested in using “The ONE Study concept” and have already put out calls (e.g. Immune Tolerance Network-ITN) which use a similar consortium principle. Furthermore, The ONE Study immune monitoring program has helped to inspire the establishment of the vGTL (virtual Global Transplantation Laboratory), which continues to develop with the purpose of improving and standardizing immune monitoring in the field of transplantation. This year (2017), a workshop was held for the vGTL in Vancover, Canada (Sponsored by The Transplantation Society) and the next workshop is planned at the upcoming Transplantation Society meeting in 2018, Madrid, Spain. The development of the vGTL has been published by The ONE Study coordinator (Transplantation 99:381, 2015).
The ONE Study group has also established a close collaboration with Prof. Angus Thomson, University of Pittsburgh. His group has been developing a donor-derived tologenic DC product to use to treat kidney transplant recipients. We have shared our experience with Prof. Thomson regarding cell manufacturing issues, have invited him to our 2016 annual ONE Study meeting, and have given him our recommendation for a clinical protocol based on The ONE Study experience. Indeed, his group now has approval from the NIH to conduct a trial that is essentially identical to The ONE Study Cell Therapy Group trials in living donor kidney transplant recipients. We remain in close contact with Prof. Thomson as they now near the time to treat the first patients, and the coordinator of The ONE Study is a member of the advisory board for this NIH funded trial. The ONE Study has had a major influence on the planned conduct of Prof. Thomson’s trial.
A close cooperation has been established with the CNTRP (Canadian National Transplant Research Program; Prof. Lori West). The CNTRP is a Canadian network of transplant centers performing research and combining efforts to improve clinical trial development. Prof. Birgit Sawitzki from The ONE Study (Berlin partner - coordinating the immune monitoring program) has helped the CNTRP (particularly, Prof. Megan Levings, Vancover), along with our associate partner Beckman Coulter, develop their immune monitoring with regard to immune phenotyping by flow cytometry. The ONE Study cooperation with Beckman Coulter has been the starting point for the development of immune cell phenotyping panels (DuroClone) that are now available globally via Beckman Coulter (see below). The concepts learned during The ONE Study have therefore been proliferated in other countries and used to a maximal advantage. This cooperation has developed extremely well, and is expected to yield an application in the Horizon 2020 program. In fact, there is a cooperative grant between the EU and Canada that may offer an opportunity to further develop our work together.
List of Websites:
ONE Study Team
Dept. of Surgery
University Hospital Regensburg
93053 Regensburg, Germany
Prof. Edward K. Geissler, PhD
Head of Experimental Surgery, Dept. of Surgery,
University Hospital Regensburg
ONE Study Trial Coordinator
Phone: +49 941 944 4895
ONE Study Project Administrator
Phone: +49 941 944 4894
ONE Study Project Manager
Phone +30 0577 50518